Original source video: NTSB [Youtube]
Transcribed with: youtubetotranscript.com
Paragraph breaks have been added manually.
Jan. 27, 2026 NTSB Board Meeting Presentations
All right, welcome to the National Transportation Safety Board.
I'm Jennifer Homundee and I'm honored to serve as chairwoman of the NTSB.
I'm joined by my fellow board members, member Michael Graham, member Todd Inman, and member Tom Chapman.
Before we begin, I'm going to yield one minute to Member Chapman.
Thank you for the opportunity to speak speak uh briefly, chair.
I'm going to excuse myself from today's proceedings as I did prior to last summer's investigative hearing.
Before doing so, I want to again explain why I will not be participating.
As I have formally disclosed in various public financial disclosure filings, my family and I continue to receive flight privileges from American Airlines pursuant to what's called an executive change in control and severance benefits agreement.
Without going into details, that's an arrangement which dates to a circumstance and time much earlier in my professional career.
American Airlines is not a formal party to the investigation.
However, I have recused myself from participate participation in this investigation because of the business and corporate relationship between American Airlines and PSA Airlines.
And PSA is a party, of course.
I commend our team and my colleagues on the board for their hard work.
I've followed the investigation closely, and I'll be observing today's board meeting by video.
A thorough determination and examination of the facts is owed to those who lost their lives in this terrible tragedy and to their loved ones who are grieving the unimaginable loss.
I know that some of the folks are here today and others are undoubtedly watching online.
My heart goes out to you.
I am confident the lessons learned from this investigation will help ensure such a tragedy can never be repeated.
and chair.
That concludes my remarks.
My thanks again for the opportunity to explain my situation.
>> Thank you, Member Chapman.
Today's meeting is open to the public in accordance with the government government in the sunshine act.
It's the first time board members and staff have come together to discuss the draft report and to consider the proposed findings, probable cause, and safety recommendations.
Before I continue, I want to pause for a safety video.
Olivia, hello and welcome to the NTSB boardroom.
Before we get started with today's meeting, I'd like to quickly walk you through our emergency procedures.
First things first, please take a moment to silence your phones or other electronic devices.
If you need the restrooms, they're just outside the boardroom on the left side of the lobby.
If someone experiences a medical emergency, our on-site federal protective service officers are trained to respond.
They'll call 911, perform CPR if needed, and use the automated external defibrill device mounted in the lobby between the boardroom exit doors.
They're also trained to provide first aid.
There's a first aid kit at the security desk in the lobby.
If we're asked to shelter in place, like during severe weather or a security incident, please stay seated and listen for instructions.
Our security team will manage the situation and let us know when it's safe to leave.
If the fire alarm sounds, please follow NTSB staff to the nearest exit.
There are two main options.
You can exit via the stairs next to the stage, or you can exit the same way you came in through the lobby.
Please walk.
Don't run.
Once outside, you'll be directed to an assembly area.
Stay there until we're clear to come back inside.
If there's a small fire, our security officers are trained to respond using one of the fire extinguishers.
You'll find them in three locations.
by the stairwells on either side of the stage at the front of the boardroom, mounted on the wall at the back of the boardroom, and near the security area in the lobby.
For anyone who needs an accessible evacuation route, you can use the door on the left side of the lobby, which leads you into the parking garage.
From there, turn right and follow the walkway to the street level.
NTSB staff will be available to assist anyone who needs help.
And finally, safety is everyone's responsibility.
If you see anything that seems off or concerning, please let a staff member or security officer know right away.
Thank you for helping us maintain a safe and secure space.
>> Thank you, Cherry, for that excellent safety v video.
And before I uh continue on with my opening statement, I'll yield two minutes to member Graham.
>> Thank you, Chair.
Today is a heavy day and I do not think that is lost on any of us.
This has been an incredibly difficult year for our agency and our investigators, the first responders to this accident and the aviation community as a whole as we have mourned the loss of 67 lives nearly one year ago from an accident and tragedy that never ever should have happened.
As difficult as this year has been for so many though, I know it has been most difficult for the family and friends who lost loved ones in this accident.
Whether you're here today with us or online or watching at a later date, please know that I, my fellow board members and our entire agency mean it when we say we have not let a day go by when we are not thinking about your loved ones.
They have been on our minds and in our hearts at every in each and every step of this process.
Our investigation has been a grueling and challenging process, but I cannot imagine having to endure all of this in the eyes of the public.
I will continue to pray for each and every one of you.
I imagine there will be some difficult moments today for all of us as we try to provide answers to how a multitude of heirs led to this tragedy.
There will be some times where individual errors may be noted throughout the course of the day in relation to this accident, but I want to make it crystal clear.
Any individual shortcomings were set up for failure by the systems around them.
We are not here today to place blame on any individual or any organization.
But we are here to ensure those systems that failed to protect 67 people on January 29th, 2025 never fail again.
To that end, I remain steadfast in my commitment to forcefully advocate for all the safety recommendations we adopt today moving forward.
Thank you, chair.
>> Thank you.
Excellent statement.
And I do want to point out this was a as I said at the uh initial press conference and all hands on deck uh investigation and onsecene work.
Member Graham, I know that you were there.
You met with families uh some of which are from your home.
So thank you very much for your work.
Uh now I want to yield uh a couple minutes to uh member Inman who was the board member on scene for this accident.
And thank you for your work.
Thanks, chair.
First off, to the staff, thank you for dedicating almost a year of your life to this.
I know for many of you, it has been a singular mission for many others they've helped and supported.
So, we appreciate it.
It is not easy what you do, nor what you face every day.
To the family members, I'm sorry on behalf of all of us.
We should not be here.
I told you yesterday, none of us should be here.
You would hope someday we could actually actually get rid of the NTSB because there was no longer a need.
But unfortunately from tragedy, we draw knowledge to improve the safety for us all.
So for that, I'm sorry.
It will not be an easy day.
And I think as Mike just said, there is no singular person to blame for this.
These were systemic issues across multiple organizations.
And I think as you'll see, we will place a lot of the information that you're going to hear on those responsible organizations and hopefully make recommendations so that this never happens again.
But while the board meeting may end today, it's not over.
Tomorrow begins the advocacy.
It begins the time to get the recommendations at times.
It's the time to actually really start the hard work for a lot of people that may go on for years.
But we will be with you throughout that time.
We're not going anywhere.
I know you're not.
We're there to work with you and to help you in any way we can.
So, for that, again, I am sorry for us being here.
I wish it did not happen.
I never want to set before another family and do the briefings like we had to do in the days after that crash.
I know we cried together and we'll probably do that again today.
Thank you.
>> Thank you, uh, member Enman.
Appreciate that.
The NTSB meets today to finalize our investigation of the mid-air collision between a Blackhawk operated by the US Army under the call sign PAT 25 and a CRJ700 operated by PSA Airlines as American Airlines Flight 5342.
The collision occurred on January 29th, 2025 at about 8:48 p.m.
Eastern Standard Time, about a half mile half a mile southeast of Ronald Reagan, Washington National Airport.
Tragically, there were no survivors.
The two pilots, two flight attendants, and 60 passengers aboard the airplane and three crew members aboard the helicopter were fatally injured.
I want to take a moment to address the 67 people, the families and friends of the 67 people who died that day.
I imagine every day since January 29th has been incredibly difficult.
And I imagine that today is even more difficult.
And Thursday, as we mark one year since this tragedy occurred, since you lost your loved ones, as a mom, I can only imagine all that you've been through.
You are in our daily prayers.
I pray for you every single day.
And on behalf of everyone at the NTSB, our entire agency, please accept our heartfelt condolences and our deepest, deepest sympathies.
Unknowingly, you are also an inspiration to so many of us.
You have in the wake of absolute devastation shown remarkable selflessness, courageously advocating for important reforms in aviation safety to save lives for others.
I have no doubt that the information uncovered by our investigation will support all of your efforts to make aviation safer, to work towards a future where no family endures such tremendous loss.
Our family assistance team and the American Red Cross are here today to provide support for anyone anyone here who might need it.
colleagues.
Uh where could you uh indicate where you are in the room right now? Uh we have several if you can ra stand raise your hand.
Uh some in the back.
If you if anyone needs support, please do not hesitate uh to ask.
We have uh private rooms that we can uh uh sit down with you and uh meet with you.
I also want to recognize NTSB's staff for your incredible work on this investigation, which is undoubtedly one of the most complex in NTSB's history.
If anyone would like a glimpse of what these dedicated safety experts have done over the past year, look no further than the public docket for this investigation.
It spans more than 19,000 pages and includes testimony from our 3-day, 33-hour investigative hearing last July.
And yet, we can never quantify the time you dedicated to this investigation over the past year.
From your work on scene to the countless hours spent analyzing and distilling the information into a comprehensive report that exceeds 500 pages to ensure every lesson is learned from this devastating tragedy.
that you did so in less than 12 months time while meeting the high bar for quality and impartiality that our agency is known for.
At one point, working through the longest government shutdown in history with a myriad of other accident investigations underway is a true testament to your dedication and your professionalism.
And we thank you for all that you do.
There is a tendency in the immediate aftermath of any investigation or any accident that we investigate to question human error, the actions or inactions of individuals.
However, human error in complex systems like our modern aviation system and the national airspace system isn't a cause, it's a consequence.
Many things need to go wrong for an accident to occur.
In any investigation, the NTSB could choose to focus on a simple moment, on what happened immediately prior to an accident, on the individuals involved.
But that's not the whole picture.
to quote research from the National Highway Traffic Safety Administration.
What we refer to as human error is in reality the last event in the causal chain immediately preceding a crash or accident.
Indeed, in board meetings over the years, in our findings and our analyses across all modes of transportation, we've often referred to the work of leading scholars like Dr.
James Rezen, Captain Dan Marino, and Professor Nancy Leon to demonstrate that human error is a symptom of deeper underlying systemic failures.
As my colleagues stated in their opening statements, a consequence, not a cause.
These underlying deficiencies, often referred to as latent conditions or systemic vulnerabilities, are what aligned to allow for the worst US aviation disaster in terms of fatalities since November 12th, 2001, when American Airlines Flight 587 crashed into a residential area of Bell Harbor, New York, killing all 260 people aboard the airplane and five on the ground.
And certainly we remember always the 50 people who lost their lives on FE February 12th, 2009 when Kolgan Airflight 3407 crashed into a residence in Clarence Center, New York.
A year before the mid-air collision at DCA, Alaska Airlines Flight 1282 experienced an in-flight separation of the left mid-exit door plug and rapid depressurization during climb after takeoff from Portland International Airport.
We were fortunate no one lost their life or was seriously injured.
But within weeks, there was a lot of focus on human error on the actions of a team of Boeing manufacturing personnel in Reon.
In the in final investigation report, we cited Dr.
James Reasonz.
Within a robust system, the introduction of a single error is almost never the only cause of an accident.
Rather, several barriers of defense must fail for the error to lead to an accident.
In other words, there was a long chain of events that led to the door plug departing the aircraft, just as there is for every accident we investigate.
In preparing for this board meeting, I reviewed a myriad, almost all midair collisions we've investigated since 1968 when a North Central Airlines Flight 261, a Convair 580 collided with a Cessna 150 near General Mitchell Airport in Milwaukee, Wisconsin.
That was one of 38 midair collisions we investigated in 1968.
A year later, we investigated the midair collision of Alageney Airlines Flight 853 and a small Piper Cherokee outside Shelbyville, Indiana, killing 83 people.
Within months, the board held a hearing on mid-air collisions in general and issued 14 recommendations aimed at preventing them from reoccurring, including our first recommendation to expeditiously develop and implement a collision avoidance system in all civil aircraft.
Fast forward 50 years to 2019.
I was the board member on scene for a mid-air collision between a dehavland DHC2 Beaver airplane and a Dehavland DHC3 Otter airplane about 8 miles northeast of Ketchacan, Alaska.
Six people died, 10 were seriously injured.
Mr.
Banning and Dr.
Bramble were also on scene.
We didn't conclude that the cause was pilot error, that one pilot hit the other's plane.
Rather, we concluded that the inherent limitations of the see and avoid concept were the concern.
The similarities between the midair collisions we investigated 50 years ago in 1969 in Ketchacan in 2019 and in DCA near DCA in 2025 are chilling in any one of them.
We could have blamed flight crews, individual pilots, maintenance personnel or controllers, but we didn't because we have long, long recognized that human error is a symptom of a system that needs to be redesigned.
That is a quote from Professor Nancy Leon.
When spaceship 2 broke up during a test flight in 2014, our probable cause didn't site human error, but scaled composits failure to consider and protect against the possibility that a single human error could result in a catastrophic hazard.
And in 2022, when Amtrak train derailed after hitting a dump truck that was blocking a grade crossing in Mendon, Missouri, we didn't blame the driver.
We found the design of the crossing was significantly flawed.
It reduced drivers ability to see approaching trains and made stopping difficult for heavy trucks.
That enabled Governor Parsons to address not only the safety of the that grade crossings, but make improvements to 49 others across Missouri, saving countless lives.
Steve Wallace, the former director of the FAA Office of Accident Investigations, was interviewed in 2023 to mark the 20th anniversary of the Space Shuttle Colombia disaster.
After the Colombia disintegrated upon re-entry to Earth, killing all seven astronauts on board, the NTSB was heavily involved in the investigation.
And in fact, one of our leading aircraft uh aerospace engineers who actually did the structural work for this investigation did the work on Colombia with NASA.
Steve was a member of the Colombia accident investigation board.
In the interview, he cites a lesson he learned from our investigators.
Quote, NTS people have a saying that when you find the human error, that's not the end of the investigation.
That's the beginning of the investigation.
What is the true root cause? The root cause is the thing that you have to change so it doesn't happen again.
Commercial aviation embraced the same shift in root cause analysis and the results were powerful.
In fact, they challenged the NTSB to do better when we only focused on human error back in the 1970s and before that.
And as a result, flying became safer.
Before this tragedy occurred on January 29th, 2025, the US aviation system was experiencing a record level of safety.
That is the power of taking a systems approach which is based not on speculation but on decades of research, evidence and our own investigations where the lessons learned turned in turn into lives saved.
As aviation safety evolves, so do the systems and so do the risks.
I'll repeat, within a robust system, the introduction of a single error is almost never the only cause of an accident.
Rather several barriers of defense must fail for the error error to lead to an accident.
We are here today to discuss those failures.
Today you will hear how deep underlying systemic failures, system flaws aligned to create the conditions that led to this devastating tragedy.
From design of our airspace around DCA to the limitations of see and avoid which we've been warning about for over five decades to failures of entire organizations to evaluate and act on readily available data heed repeated recommendations and foster robust safety cultures to name a few.
Months ago, I received a letter from someone sitting in this audience today whose fiance was aboard 5342.
In that letter, he challenged us to quote, "Leave no stone unturned, ask the hard, uncomfortable questions that will ruffle feathers and let no one obfuscate or delay the truth.
I assure you, we did just that.
You will see today that we left no stone unturned.
We asked the hard, uncomfortable questions that ruffled feathers and we got to the truth.
Our work, however, doesn't end today, as member Enman stated with the issuance of a final report.
This is just the first step.
We must, we must, it is our duty to relentlessly, vigorously pursue safety change, which means we must do everything in our power, even ruffle feathers, to get our recommendations implemented or this will happen again.
I'd like to thank the first responders on behalf of the board.
I have a slide listing the all of the first responders that were on scene.
The A8 organizations tremendous work.
As you can see, dozens of organizations responded to the scene of this tragedy to support search and recovery efforts and assist in our investigative work.
law enforcement, fire officials, dive teams, federal, state, and local agencies.
We don't often talk about, but they're human, too.
And what they experience can also be quite difficult.
So, thank you if you're listening or you're here.
Thank you for your public service, for your dedication, and for your heart.
You embody the best.
You embody the best, each of you, of what public service means.
Now, let's turn to today's meeting.
We're going to do things a little bit differently today as there's a lot of complex information that we have to go through.
We've divided staff presentations into three sections, each of which will be followed immediately by board member questions.
We believe this is the best way for those watching and in the room to follow along.
So, let me go through today's agenda.
Media will want to get their press their their little pens out.
You'll know.
Look, they are all actually doing it right now.
Um, okay.
So, in a moment, uh, Brian Curtis, our de deputy managing director for, uh, investigations, will introduce the staff.
Then we have 75 minutes of staff presentations.
Uh starting with our investigator in charge.
Then we will move to air traffic control.
We will discuss air traffic control and human performance.
And then we will go through air traffic control post accident drug and alcohol testing.
We will then have 45 minutes or so of board member questions and uh then we will take a break a 15minut break which should occur somewhere around 11:30.
We will then move into a second group of staff uh presentations that will take about 45 minutes.
We'll start with collision avoidance and ADSB, move into helicopter operational factors, and then move into human performance.
Then there will be 45 minutes or so of board member questions just on that section and relevant material in our docket and the draft board report.
Then we will break for lunch.
I imagine assuming everything goes right and I mean that we don't lose heat or video.
uh that I've I've probably doomed us now.
Um uh are we'll go for a lunch break that should occur shortly after 1:00.
Uh we'll have an hour for lunch.
I want to make sure that our staff most importantly has the break they need to eat, to get water or Gatorade, and uh to regroup before the next portion of the board meeting.
So after lunch, we'll come back around a little bit after 2:00.
Uh then we'll go through staff presentations focused on safety management systems, safety culture, and safety data.
After that, we'll have board member questions focused just on that portion uh on that presentation and relevant material in the docket and the board report.
Then we'll go to a 15minute break that should be slightly after 3:00 and then we'll go through the proposed findings, the proposed probable cause and the proposed recommendations as well as the draft report.
Uh Mr.
Curtis will be supplied with a lot of water because he will do be doing a lot of reading from what I understand.
His reading alone on the findings, probable cause, and recommendations in total will take about an hour and 10 minutes.
That's not even considering the time that we'll we'll have for deliberation.
We have numerous findings and numerous recommendations.
And after that, I'll have a brief o closing statement and then we'll adjourn.
Okay.
So, I'll now turn it over to Mr.
Curtis, our deputy managing director for investigations to introduce the NTSB team.
>> Thank you.
Good morning, chair, and good morning to the board.
I'll now introduce staff for today's meeting.
Unless otherwise noted, staff are from the Office of Aviation Safety.
In the front row to my right, Tim Learn, director for the Office of Aviation Safety.
Bryce Banning, investigator in charge for this investigation.
Brian Soaper, air traffic control.
Dr.
Katherine Wilson, air traffic control and human performance.
Dr.
Tran Kaagill, medical from the office of research and engineering.
Behind me in the second row, left to right, Jennifer Adler, director for the office of safety recommendations and communications.
Tom McMurray, general counsel for the NTSB.
Carl Schulist, vehicle performance chief from the office of research and engineering.
Allison Diaz, report writer, Sarah Lewis, air traffic control, and Marie Mohler, aircraft performance from the office of research and engineering.
And in the third row, left to right, Olivia Fowler, audiovisisual for today's meeting.
Chun Shen, helicopter operations.
Mike Richards, meteorology.
John O' Callahan, aircraft performance from the office of research and engineering.
And Nathaniel Hoy, recommendations from the office of re safety recommendations and communications.
Additional staff present for to answer questions from the board today.
Captain Van McKenna, helicopter operations.
Dr.
Jenna Price, safety data from the Office of Research and Engineering.
Dr.
Bill Bramble, human performance.
Michael Portman, Recorders from the Office of Research and Engineering.
Caleb Wagner, systems.
Captain Rocky Stone, airplane operations.
And Dr.
Lauren Grath, safety data from from the office of the managing director.
The presentations will begin now with investigation overview by the investigator in charge, Bryce Banning.
Mr.
Banning.
Good morning, Chairwoman Hammedy, member Graham, and member Iman.
My presentation will begin with an animation covering history of flight, helicopter routes, and an overview of the air traffic control communications shortly before the accident.
Then I will discuss the visibility study and play a second animation which will depict simulated views from the cockpits of each of the accident aircraft.
I will also cover the wreckage examination, exclusionary findings, safety issues, and previously issued safety recommendations.
Air traffic control audio is included in these animations.
Cockpit voice recorder audio from both airplane and a helicopter are not included in these animations, but their transcript will be shown.
We realize that some attendees and remote viewers may find these animations difficult to view and hear.
We will pause before starting the animations to allow those currently in the boardroom to leave for the duration of the animation or turn away from your screens if viewing remotely.
We will inform you when the animations are finished playing.
We will begin the first animation now and we'll pause for those who would like to leave the room or turn away from their screens.
The animation will play for approximately 12 minutes.
We will be dimming the lights in the boardroom so that people can see the animation better.
On January 29th, 2025, about 8:48 p.m.
Eastern Standard Time.
A Sakorski UH60 Lima helicopter operated by the United States Army and a PSA Airlines CRJ700 collided in flight approximately 1 half mile southeast of Ronald Reagan Washington National Airport or DCA.
The two pilots, two flight attendants, and 60 passengers on board the airplane, and all three crew members on board the helicopter were fatally injured.
The military helicopter flight originated from Davidson Army Airfield, Fort Belvore, Virginia.
The flight was an annual evaluation for the pilot with the use of night vision goggles.
The airplane was operated by PSA Airlines doing business as American Airlines flight 5342, a scheduled domestic passenger flight under title 14 of the code of federal regulations part 121 from Witchah, Kansas to Washington DC.
Night visual meteorological conditions prevailed at the time of the accident.
In the radio communications, the accident airplane was using call sign Blue Streak 5342 and the Army helicopter was using call sign PAT25.
The Federal Aviation Administration provides charts that specify helicopter routes to facilitate operations in hightra areas.
The helicopter routes have no defined lateral boundaries, but they do have maximum altitude limits.
The helicopter routes are depicted by a blue line on the Google Earth image.
Helicopter Route One's western section begins in Cabin John, Maryland, and follows the Ptoac River southeast to Washington DC.
Through the Tidal Basin and the Washington Channel to Hannes Point and then turns east along the Anacostia River.
At the time of the accident, helicopter route 4 intercepted Route One at Hannes Point and continued south along the Ptoac Rivers eastern shore past DCA and the Wilson Bridge.
North of Chainbridge, the maximum altitude is 1,300 ft.
Between chainbridge and keybridge, the maximum altitude is 700 ft.
Between Keybridge and Memorial Bridge, the maximum altitude is 300 ft.
And south of Memorial Bridge to the Wilson Bridge, the maximum altitude is 200 ft.
At 8:33 p.m., 15 minutes before the collision, Pat 25 requested permission to travel from Cabin John along helicopter route one to helicopter route 4 to Davidson Army Airfield.
The DCA local controller approved the request.
>> Kevin route one route forward.
>> 25 approved.
>> There was no communication between PAT 25 and the DCA local controller for 12 minutes.
At 8:39 p.m., 9 minutes before the collision, flight 5342 was cleared for the Mount Vernon visual runway one approach to DCA by PTOAC terminal radar approach control.
Select incockpit communications between the instructor and the pilot in PAT 25 primarily about altitude will be shown on a black background in the animation.
At 8:43 p.m.
5 minutes before the collision, flight 5342 was handed off to the DCA tower local controller.
The local controller asks the crew of flight 5342 if they could switch their approach from runway 1 to runway 33.
>> 5342 Washington 320 7 + 25.
You take runway 33.
>> The crew accepted the change and they were cleared to land on runway 33.
This magnified satellite photograph shows the locations of runway one and runway 33 at DCA.
>> Yeah, we can do 332 at the Wilson runway 33.
Runway 33 clearway.
There was no further communication between the local controller and the crew of flight 5342 before the collision.
This animation shows aircraft in the vicinity of DCA for the 3 minutes before the collision.
The local control and a helicopter control positions were combined and the local controller was communicating with six airplanes and five helicopters.
Communications took place on an airplane radio frequency and a helicopter radio frequency.
The local controller was broadcasting on both the airplane and helicopter radio frequencies, but airplane communications were only heard in other airplanes and helicopter communications were only heard in other helicopters.
The animation includes air traffic control audio for both the airplane and helicopter frequencies.
A transcript of the controller communications will appear in the upper left and the local controller will be indicated on the map.
A transcript of aircraft communications will appear in the lower left and the aircraft call sign will be highlighted on the map.
The controller was also communicating with aircraft outside the bounds of this map which will be indicated.
The accident helicopter is shown in orange with the label PAT 25 and the accident airplane is shown in magenta with the label PSA 5342.
J 78 9 1 Z 320 Z5 gust 25 no delay runway one clear take off takeoff runway one no delay J 7 8 9 one ro for exchange for good night roger >> Washington American 3130 with runway one >> American 30 to130 Washington 3104 go to five from one clear to land Check for depart may depart Bravo.
Okay, clear land runway one.
>> Yes.
789 contact.
>> Okay.
Come back 3130 for the final >> 25 traffic just south of Wilson CR at 1,200 ft.
Start from runway 33.
American 1630 tower runway oneway 33 runway one.
American 1630 standby.
53 07 keep roll out November ground 47.
Good day.
November 37.
>> American 16 320 14 go 25 mile left base runway 33.
No delay.
Runway one clear to take off.
Number one, clear for takeoff.
American 1630 m 7 request.
You say rout 7 approved evening helicopter 100 ft 2 ner 91 on the altimeter requesting zone 6 air one alter 2 ner n zero approved through class space for legac 29 copy 25 in flight passj.
American 472 one star 320.
The following video shows the airplane and helicopter leading up to the point of collision.
The video was recorded by a camera located near the approach end of runway 33 at DCA oriented to the south.
The video shows the night lighting conditions and the appearance of the aircraft lights against the cultural lighting of the city.
Aircraft lined up on approach to DCA can be seen in the right corner of the screen.
The flashing lights identify the touchdown zone of runway 33's landing threshold.
Flight 5342 is visible in the upper left corner as it approaches runway 33.
Pat 25 appears from the left side of the screen moving to the right as the two aircraft cross flight paths.
The animation is done playing.
A visibility study documents the location, size, and movement of other aircraft in each respective pilot's field of view and how the aircraft structure might limit that field of view.
Human performance investigators can determine how conspicuous each aircraft would be to the flight crew and how that impacts their see and avoid task.
The visibility study faced additional challenges due to the night conditions and use of night vision goggles by the helicopter crew.
As we will discuss, it is difficult to accurately represent how the night lighting or cultural lighting looked to the crews at the time of the accident.
Investigators documented the cultural lighting around DCA in a series of night observation flights conducted on March 27th and 28th of 2025.
We chose these observation dates to replicate the conditions on the night of the accident.
Crews flew the accident flight paths individually using exemplar aircraft which were observed by human performance operations and air traffic control investigators.
We recorded the view in both videos and photos from each aircraft.
the air traffic control tower and on the ground.
This documentation was used to determine which available flight simulator program best represented the cultural lighting at the time of the accident and inform the human performance analysis.
Shown is a video frame recorded from the cockpit of the CRJ night observation flight when turning on to final for runway 33.
This captures a view of the cultural lighting in the DC area.
However, no device records exactly how the human eye sees a scene.
These types of images from various recording devices and the impressions of the human performance operations and air traffic control investigators from the night observation flights were used together to inform the analysis of the difficulty of the see and avoid task.
The visibility study recreates the view from the cockpit and includes what the airplane structure would obscure.
The photo on the left is of a CRJ700 cockpit and labels the four cockpit windows, left window, left windshield, right windshield, and right window, and indicates the location of the captain, who was the pilot flying, and the first officer, who was the pilot monitoring.
The image on the right shows the field of view of the captain looking forward through the left windshield.
The light gray mask indicates how the structure of the airplane would have obscured some of the view.
The night imagery is from Microsoft Flight Simulator 2024 and the simulated view shown is from 35 seconds before the collision.
The location of PAT 25 is circled, but it is difficult to detect and identify against the cultural lighting of Washington DC.
The left photo shows the helicopter cockpit and labels the left window, left windshield, center windshield, right windshield and right window and indicates the location of the pilot who was the pilot flying and the instructor pilot who was the pilot monitoring.
The image on the right shows a view of the instructor pilot in the right seat looking forward.
In addition to the aircraft structure, the image on the right includes a mask of the night vision goggle view looking straight ahead.
While the MVGs are binocular, the brain merges the image into a single circular view.
In this image, the MVG view is shown as white phosphor, but the animation will show a green phosphor display.
Both types of MVGs were available to the crew.
This image is from 19 seconds before the collision, and the location of the CRJ is circled.
While the location, size, and movement of each aircraft in a pilot's field of view is accurate, there are important limitations to the visibility study, including first, the accuracy of the cultural lighting.
You will see in the animation that some landmarks are unrealistically bright.
Additionally, not all of the aircraft south of the Wilson Bridge were visible in the simulation.
Secondly, the accuracy of the external lighting on each aircraft, which affects how visible it is against the cultural lighting.
The CRJ700 model for Microsoft Fi Flight Simulator 2024 did not have functional external lighting.
So, a model of a Boeing 737 was used instead.
The differences in the appearance of these airplanes at night is insignificant until just before the collision.
Lastly, the accuracy of the night vision goggle view.
The goggles are shown stationary and the pilot's head does not move or scan.
The imagery also does not include additional light sources that night vision goggles would show more brightly, such as stars.
We will begin the second animation now and we'll pause for those who would like to leave the room or turn away from their screens.
The animation will play for approximately 8 minutes.
We will be dimming the lights in the boardroom so that people can see the animation better.
This animation presents a simulated view from the right seat of the helicopter.
A dark gray structural mask generated from a three-dimensional laser scan of an exemplar aircraft is overlaid on the scene.
A light gray mask depicts the mounting structure of the night vision goggles positioned for the instructor pilot looking straight ahead.
The green tint mimics a green phosphor visual effect, but the actual behavior of the night vision goggles has not been simulated.
Crew communications transcribed from the cockpit voice recorder are shown as text at lower left.
Audio recordings of air traffic control communications on the helicopter frequency are synchronized with the animation with corresponding transcript text displayed in the center.
Radio interference blocked some transmissions from the local controller and other helicopters.
Transmissions that were blocked are indicated by striking through words in the transcript.
A moving map depicting the locations of PAT 25 and flight 5342 and synchronized with the view from the cockpit will be shown at lower right.
78 91 star 3205 gust 25 travel 2 no delay runway one clip takeoff >> flag one >> flag one roger changeburg good night roger American 301 130 watching tower one 310 14 go five from wave one clear to land check The department may depart parts Yeah.
at 789 contact American 3130 slow 25 traffic just south of Wilson bridge 0 at 1,200 ft runway 33 American 1630 tower runway one boy traffic runway 33 Dischar stand by 53 07 keep roll out November ground 47.
over there.
Good evening.
>> American 16 320 14 mile base runway 33.
No delay.
Runway 1 clear to take off.
Most 7 request you say router.
Good evening.
Air 1 single back helicopter to 100 ft 2 ner niner one the altimeter requesting zone 6 for max air one alometer 2 ner ner zero approved for class to fair two ner zj in flight 25 passed Aircraft separation separ.
>> This animation presents a simulated view from the right seat of the CRJ.
A dark gray structural mask generated from a three-dimensional laser scan of an exemplar aircraft is overlaid on the scene.
Crew communications transcribed from the cockpit voice recorder are shown as text at lower left.
Audio recordings of air traffic control communications on the airplane frequency are synchronized with the animation with corresponding transcript text displayed at lower right.
A moving map depicting the locations of both aircraft synchronized with the view from the cockpit is shown at upper left.
J 789 one star 3205 go to traffic 2 no delay runway one clear take off takeoff runway one no delay 789 one ro for change for good night pass two roger >> tower American 3130 with you runway one >> American 3130 tower 1 3 1 04 go to five runway one clear to land Check when depart may depart part.
Yeah.
Clear to land runway one.
>> Yes.
789 contact.
>> Okay.
American 3130 slow 25 traffic just south of Wilson bridge CRJ at 1,200 ft runway 33 American 1630 tower runway one boy traffic runway 33 dis runway one American 1630 by one standby 53 07.
Keep roll out.
November ground 7.
Good day.
November 37.
American 16 to one.
Star 320 4.
Go to five.
Traff 33.
No delay.
Runway one.
Clear for B to take off.
>> Number one, clear for takeoff.
American 1630.
Most 7 request.
Most of you say route three zone six seven approved med air one Washington two ner n zero approved for class rover space for land to Back at 25 CRJ in flight 25 pass 472.
American 472 Washington 320 7.
The animation is done playing.
The wreckage of both aircraft were located in the Ptoac River in water depths ranging between 1 to 8 ft depending on location and title conditions.
About 90% of the structure of both aircraft were recovered.
Examination of the recovered wreckage indicated that the helicopter's main rotor contacted the underside of the airplane's left wing, resulting in the outboard section of the wing separating in flight.
Video evidence showed both aircraft rapidly descending and impacting the water after the collision.
The NTSB's investigation's comprehensive review of the accident circumstances determined that the following did not contribute to the cause of the accident.
The pilots of flight 5342 were certificated and qualified in accordance with federal regulations.
The pilots of flight 5342 were medically qualified for duty and available evidence does not indicate that they were impaired by effects of medical conditions or substances at the time of the accident.
Review of the pilot's time since waking and sleep opportunities in the days before the accident indicated that the pilots had adequate rest opportunities.
The pilot instructor pilot and crew chief on board PAT 25 were qualified and current in their positions as designated by the unit commander in accordance with Army regulations.
The pilot, instructor, pilot, and crew chief were medically qualified for duty, and available evidence does not indicate that they were impaired by effects of medical conditions or substances at the time of the accident.
Review of the three crew members time since waking and sleep opportunities in the days before the accident indicated that the crew had adequate rest opportunities.
Both aircraft were properly certificated, equipped, and maintained.
the operations supervisor and four controllers who were working in the DCA air traffic control tower cab at the time of the accident were certificated and current in accordance with federal regulations.
The local controller, assistant local controller, and operation supervisor were medically qualified for duty, and available evidence does not indicate that they were impaired by effects of medical conditions at the time of the accident.
Review of the local controller, assistant local controller, and operations supervisors time since waking and sleep opportunities in the days before the accident indicated that the controllers had adequate rest opportunities.
A review of observations recorded throughout the night of the accident revealed no evidence of any local atmospheric pressure anomalies that would have impacted barometric altimeter readings.
The investigation explored multiple safety issues.
the design of the Washington DC area helicopter routes and operators understanding of the route structure and limitations.
Air traffic controller workload position combining and communications practices.
the extensive use of pilot applied visual separation and the inherent limitations of see and avoid deficiencies in FAA safety culture and postac drug and alcohol testing the limitations of the traffic awareness and alerting systems on both both aircraft and shortcomings in FAA and US Army safety assurance and risk management processes, including lack of proactive data sharing and analysis to identify and mitigate midair collision risk.
Early in the investigation, the NTSB found insufficient separation between helicopter traffic on Route 4 and aircraft landing on runway 33 and recognized an alternate route for helicopters was needed to prevent a mid-air collision.
As a result, the NTSB issued two urgent SEF safety recommendations on March 11th, 2025.
The first recommendation asked to prohibit operations on helicopter route 4 between Hannes Point and the Wilson Bridge when runways 15 and 33 were being used at DCA.
The second recommendation asked to designate an alternative helicopter route that could be used to facilitate travel between Hannes Point and the Wilson Bridge when that segment of Route 4 was closed.
Immediately following the accident, the FAA implemented temporary airspace restrictions around DCA.
On March 14th, 2025, the FAA removed from helicopter route charts the section of helicopter route 4 between Hannes Point and the Wilson Bridge.
Additionally, the FAA prohibited use of DCA runways 1533 and 422 during specific limited helicopter operations in the vicinity of DCA.
What we found about helicopter route design was that airplane pilots were unaware of the possible conflict between helicopter routes and approach and departure corridors at DCA.
Pilots were not aware of these conflicts because the current aeronautical charts do not provide adequate information.
This is not unique to DCA.
Lastly, the FAA was unable to provide documentation of the required annual reviews for the Baltimore Washington helicopter route chart.
As a result, staff has proposed five recommendations to the FAA in this area.
The following three slides list the names of the NTSB staff that supported the accident investigation.
NTSB invited qualified parties to participate in our investigation.
The next two slides list out the parties.
In accordance with the provisions of annex 13 to the convention on civil aviation, the transportation safety board of Canada representing the state of design and manufacturer of the airplane provided an accredited representative.
Technical advisors from Transport Canada and MHI RJ Aviation participated in the investigation.
Staff will be giving presentations on the following topics.
air traffic control, human performance as it relates to air traffic control, and the FAA's post accident drug and alcohol testing of air traffic controllers.
A discussion will follow these presentations.
Then presentations will be provided for automatic dependent surveillance broadcast or ADSB and collision avoidance systems, helicopter operations and human performance as it relates to both aircraft.
A discussion will follow these presentations.
Lastly, we will have a presentation on safety data and safety management systems followed by discussion rounds.
Chairwoman Homy, this concludes my presentation.
Mr.
Soer will now discuss air traffic control.
>> Good morning.
My presentation will discuss the airspace surrounding DCA helicopter route 4 formation and activities of the DCA ATC tower helicopter working group.
The DCA tower facility level classification, traffic management challenges at DCA, combining of the local and helicopter control positions, use of visual separation, traffic advisories, ATC safety alerts, and positive control.
The airspace surrounding DCA is uniquely complex being a class B airspace that includes the Washington DC special flight rules area or CIFRA, the flight restricted zone or freeze as well as prohibited areas.
The tower's delegated airspace at DCA is also more expansive than that seen at similar airports, covering more than 520 square miles and containing extensive helicopter operating areas and routes charted on the Baltimore Washington helicopter route chart.
Helicopter Route 4 was established in 1986 in the first edition of the Baltimore Washington helicopter route chart and had changed very little since that time.
Helicopter Route 4 was one of several charted routes designed for the transition of helicopters through the DCA class B airspace.
The route 4 design had always posed a pro a proximal conflict to the runway 33 approach corridor and runway 15 departure corridor but it would continued to be used through the years.
The routes were never procedurally separated meaning there were no written procedures in place to ensure that when applied aircraft would be automatically separated by a required minimum.
While we learned from the FAA during the investigative hearing that these routes were never constructed to provide or ensure separation, we also learned that there was a lack of clear and consistent understanding and varied interpretations across user groups.
The DCA ATC tower helicopter working group was formed as a direct result of a significant midair near midair collision in 2013 between a helicopter and an airplane in the very same location of this accident.
The working group was composed of ATC personnel from DCA Tower only with local helicopter operators being invited as necessary to address identified safety concerns as they arose.
The working group proposed several recommendations to the FAA, including one to remove or relocate Route 4 in order to deconlict with the runway 33 approach corridor and runway 15 departure corridor.
as well as the addition of hotspots to aeronautical charting to bring awareness to pilots about the increased risk potential when operating in these areas of the proximity events.
The FAA chose not to act on the working group's recommendation to remove or relocate route 4.
Additionally, the FAA rejected the working group's recommendation to add hotspots to helicopter route charts.
In 2018, due to a change in the method in computing facility level traffic counts, DCA tower was downgraded from a level 10 to a level 9 facility.
The FAA declined to provide detailed criteria or the metrics used to establish DCA's facility level, but information available suggested that total traffic count was the driving factor in determining facility levels.
In general, staff feel that there was an inadequate consideration given to airspace, airport, and operations complexity in the downgrading of DCA's tower facility level and negatively impacted facility health as they experienced an inability to retain or recruit the experience level needed in control staff.
In addition, no consideration appeared to be given to the high cost of living in the DC area with a 15% difference in pay between levels 9 and 10, the biggest salary gap between any two air traffic control facility levels.
We found that DCA Tower had significant airfield, airspace, and mixed fleet operations complexities that were inconsistent with their current facility level.
We also found that aside from the traffic count alone, FAA provided no evidence that operational complexities nor cost of living were considered in their criteria for establishing air traffic control facility levels.
staff have proposed two recommendations to the FAA in this area.
According to public records, as of 2023 and reportedly still accurate today, DCA had the busiest single runway in the National Airspace System.
Total operations exceeding those documented at airports including Chicago O'Hare, Los Angeles, Dallas Fort Worth, and LaGuardia.
DCA was a capacity constrained airport and their traffic flow was largely managed through airport arrival rates, miles in trail spacing and slot controls.
Making this management more difficult was the limited surface area and complex airfield geometry at DCA that required regular workarounds by ATC in order to avoid delays or gridlock.
The FAA used slots to limit scheduled traffic at capacity constrained airports like DCA as well as John F.
Kennedy International Airport and LaGuardia International Airport in New York.
DCA was limited to a maximum of 67 slots per hour, 60 of those by regulation and an additional seven allowed by statute.
According to postac interviews conducted with DCA ATC staff, certain air carriers would routinely front and backload their scheduling into the first half hour and last half hour of a 2-hour block, which resulted in periods of compacted demand.
This resulted in DCA reaching total operations of 80 or more per hour at times.
DCA slots are allocated in 60-minute blocks.
After LaGuardia experienced similar problems, a rulemaking change was adopted that requires their slots to be allocated in 30 minute blocks, providing a more consistent traffic flow.
Timebased flow management or TBFM was a system whose core function was to provide the ability to schedule aircraft within a stream of traffic to reach a defined constraint point at specified times creating a timeordered sequence of traffic.
Though reportedly equipment had been installed and the capability was there for several years, the FAA had not activated the system for DCA.
Additionally, American Airlines testified at the investigative hearing that TBFM was used at several of their other hub airports and that it smoothed out the volume of traffic while also providing more accurate miles and trail.
We found that DCA Tower routinely received less than the required miles in trail spacing from the PTOIC Tron, requiring controllers to build spacing to accommodate departures using various methods.
A common mitigation strategy was to periodically offload arrivals to runway 33 as they were doing on the night of the accident.
This involved requesting crews that were on approach to land on runway 1 to change their landing runway and circle to land on runway 33 instead.
We found that DCA would benefit from established rulemaking similar to that used by LaGuardia requiring allocation of traffic in 30 minute blocks rather than 60-minute blocks.
We also found that timebased flow management would provide DCA tower with a consistent flow of traffic, more accurate spacing, and greater predictability.
We recently learned that phase one of this went operational in October of 2025 with full implementation planned to be completed by March 2026.
Staff have proposed three recommendations to the FAA in this area.
Combining of the control positions is common place in air traffic control and in facilities like DCA tower.
Detailed policy guidance is required to reduce the risk by preventing the combining the combining of positions under the wrong circumstances.
The policy at DCA regarding staffing of the helicopter control position had changed several times over the years, becoming less restrictive each time.
During prescribed hours, the policy went from helicopter control must be open in 2016 and only authorized to be combined with air traffic manager or support manager approval to 2023 becoming helicopter control must normally be decombined meaning staffed separately.
the 2024 then in 2024 it became helicopter control should normally be decombined and it authorized the positions to be combined at the supervisor or controller in charge level's discretion.
The softening in this policy from helicopter control must be open to helicopter control should normally be decombined with no requirement of a corresponding log entry or rationale open the policy up for varied interpretation.
On the day of the accident, the helicopter position had been open or staffed separately for less than one and a half hours the entire day and had been combined with local control for more than 5 hours leading up to the time of the accident.
There was no requirement to log the combining of helicopter and control positions, nor any position combination at DCA tower.
Staff have proposed one recommendation to the FAA in this area.
We will now show an animation and will pause for those who would like to leave the room or turn away from their screens.
The animation will play for approximately 3 minutes.
We will be dimming the lights in the boardroom so that people can see the animation better.
This photograph from the interior of the DCA air traffic control tower shows the view to the southeast across the Ptoic River.
Traffic displays can be seen overhead.
The traffic displays can be moved to suit the needs of the staff in the tower.
This animation presents a simulated view from the tower.
The point of view was based on the local controller's recollection of his position.
The local controller could have changed position or orientation slightly on the night of the accident.
The underlying animation was produced by the Federal Aviation Administration target generation facility simulation branch.
A dark gray structural mask generated from a three-dimensional laser scan of the tower conducted by NTSB staff is overlaid on the scene.
Audio recordings of air traffic control communications are synchronized with the animation.
Local controller broadcasts are shown in text at lower left.
Aircraft broadcasts are shown in text at lower right.
American 1630 14 travel 2 mile left base runway 33 no delay runway one clear to take off one clear for takeoff.
American 1630 >> 7 request >> you say route three zone six >> seven approved fer 91 requesting zone six bank air one watching Just alter 2 ner n0 approved through class transport shack 2 ny 25j in flight 25 pass >> aircraft 472 my Ask visual >> America 472 Washington 1 star 207.
>> The animation has completed playing.
There are two types of visual separation.
Tower applied, where the tower controller sees the involved aircraft and issues positive control instructions to affect separation.
And pilot applied, where the pilot of one aircraft sees and maintains visual contact with the other aircraft and provides their own separation by maneuvering as necessary to avoid it.
Visual separation was heavily relied upon and was the primary means of separating helicopter and fixedwing traffic in the DCA area when weather conditions permitted to not increase controller workload, traffic congestion, and traffic complexity.
Controllers were motivated to and routinely authorized visual separation for helicopters transiting the DCA airspace and helicopter crews were encouraged to request visual separation as early as possible.
According to facility records, DCA Tower had a documented history of controllers lacking in the understanding of basic fundamentals in pilot applied visual separation as well as a history of the applica of the mislication of pilot applied visual separation.
The local controllers's initial traffic advisory provided to PAT 25 reference PSA flight 5342 was technically compliant, but no follow-on or updated traffic advisories were provided.
There was no corresponding traffic advisory provided to PSA flight 5342 at any time.
The conflict alert first activated between PAT 25 and flight 5342 seconds before the collision while they were still about 1.6 mi apart and on converging courses.
Considering the proximity of both aircraft and their continually converging courses, according to the radar replay of the event, targets were such that merging target procedures should have been provided, which would have included traffic advisories to both aircraft.
The local controller stated they were concerned about both the close proximity of PAT 25 to PSA flight 5342 and their converging courses.
This coupled with the conflict alert that was active at the time.
The controllers should have issued a safety alert which would have included updated traffic advisory information and an alternate course of action if feasible.
Neither were done in this case.
Had a had a safety alert been issued, it would have increased the situation awareness of both crews and alerted them of their closing proximity to one another.
Additionally, a timely safety alert may have allowed action to be taken by one or both crews to avo.
Positive control is defined by the FAA as the separation of all air traffic within designated airspace by air traffic control.
And other FAA guidance included elements of positive control as taking command of control situations and not acting in a hesitant or unsure manner.
And that a controller exercising positive control observes present and considers forecasted traffic to produ to predict task overload and takes appropriate action to prevent or lessen the situation.
Positive control was not exercised at all times on the night of the accident.
Had the local controller exercised appropriate positive control with timely safety alert on the evening of the accident, one or both flight crews may have been able to take action in time to the avert the collision.
Thank you.
Dr.
Wilson will now continue with her presentation.
>> Good morning.
I will discuss DCA air traffic control tower use of visual separation on the night of the accident.
The concept of situation awareness, combining the local control and helicopter control positions, recognition of impending collision and expectationdriven process, the conflict alert system, radio communications when the local control and helicopter control positions are combined, time on position limitations, threaten air management, and risk management.
Given the proximity of helicopter routes and zones to the approach and departure corridors for fixed wing traffic, applying pilot applied visual separation reduced controller workload and airspace congestion and complexity.
As such, DCA ATC tower controllers were motivated to provide a traffic advisory and authorized visual separation for helicopters transiting DC airspace as early as possible.
The local controller's expectation that PAT25 crew had flight 5342 in sight was not necessarily valid.
As a result, there was potential for the hel the local controller to overestimate the level of traffic awareness the PAT25 crew had and to underestimate the level of information and assistance they might subsequently require to ensure collision avoidance.
Situation awareness forms a basis for decision-making.
Situation awareness is not only what the controller is perceiving in the current air traffic situation, but how they interpret that information and use it to project the future state of traffic moving in their airspace.
Controllers must maintain awareness of each aircraft they are managing to include, for example, location, altitude, and air speed and anticipate where the aircraft will be in the seconds and minutes to follow.
A controller's ability to maintain situation awareness is impacted by task and environmental factors such as their workload and divided attention and individual factors such as training and experience.
As cognitive capacity is reduced, for example, with increasing workload due to increasing traffic complexity, traffic volume, and/or frequent radio communications, a controller's ability to maintain situation awareness is reduced.
Because the local control and helicopter control positions were combined, the local controller was responsible for coordinating arrivals and departures, as well as VFR helicopter traffic transiting DCA airspace.
In a post accident interview, the local controller stated that he felt a little overwhelmed about 10 to 15 minutes before the accident when the traffic volume increased to 10 aircraft consisting of five airplanes and five helicopters.
But he felt the volume was manageable when one or two helicopters left the airspace.
About 90 seconds before the collision, the traffic volume increased to a maximum of 12 aircraft consisting of seven airplanes and five helicopters.
Radio communication showed that the local controller was shifting his focus between airborne, ground, and transiting aircraft.
Keeping the local control and helicopter control positions combined on the night of the accident increased the local controllers's workload and reduced his situation awareness.
Controllers routinely monitor the current state of an aircraft and predict its future location in relation to other aircraft.
Conflicts that develop slowly, particularly at night, are inherently difficult for people to recognize due to reduced visual cues.
Gradual change can reduce situation awareness and delay recognition.
The local controller expected that PAT25 would remain clear of flight 5342 because the PAT 25 instructor pilot stated that he had the airplane in sight and would maintain visual separation.
The frequent use of pilot applied visual separation reinforces the expectation that the pilot of one aircraft will maintain separation from another aircraft.
Because it is repeatedly worked as expected, it can be more difficult for a controller to notice deviations, especially when workload is high, like on the night of the accident.
The conflict alert system acts as a safety net designed to draw a controller's attention to a potential conflict using an oral alert, a flashing red CA on the certified tower radar display or CTRD, and a conflict list on the CTRD indicating in red the aircraft involved.
The image on the right depicts the red conflict alert that appeared for the accident aircraft.
Conflict alerts are presented the same regardless of the severity of the conflict.
Conflict alert activation is common at DCA ATC tower.
NTSB staff are concerned that frequent nuisance alerts may desensitize controllers to the alerts and lead to slower reaction time.
Providing controllers with additional salient cues regarding the severity of a potential conflict would reduce controller cognitive load and likely improve response time.
Staff have proposed two recommendations to the FAA in this area.
The DCA ATC tower utilized a discrete frequency for communicating with helicopters to avoid interference and frequency congestion even when the local control and helicopter control positions were combined.
Separate frequencies also made the process of decombining the helicopter and local control positions easier.
When the local control and helicopter control positions were combined, airplane and helicopter pilots could hear all transmissions made by the controller.
However, transmissions made by helicopters were not audible to airplanes, and transmissions made from airplanes were not audible to helicopters.
This alleviated frequency congestion.
However, pilots reported that hearing all transmissions would improve their situation awareness.
Staff have proposed one recommendation to the FAA in this area.
The operation supervisor had been working on a control position for over four hours at the time of the accident and had been working on the operation supervisor position specifically for the 2 hours before the accident.
Given his extended time on position, it is likely that the operation supervisor was experiencing reduced alertness at the time of the accident, which decreased his ability to effectively assess operational risks.
The collective bargaining agreement between the National Air Traffic Controllers Association or NATKA and the FAA require relief opportunities for all operational staff.
However, the agreement does not cover supervisory personnel and there are no mandatory relief periods for supervisors.
Staff have proposed one recommendation to the FAA in this area.
Threat and error management originated in the human factors literature and was first applied in flight tech operations before being applied in air traffic control.
Threaten error management or TEM is a process for identifying safety risks in the environment, analyzing those risks, and minimizing or mitigating those risks.
In November 2016, the NTSB issued safety recommendation A-16-51 asking the FAA to provide initial and recurrent training for air traffic controllers on controller judgment, vigilance, and/or safety awareness with specific reference to two midair collisions that occurred in 2015.
In July 2017, the FAA responded that instruction on threat and error management was being delivered to controllers as part of 2017 instructor-led recurrent training and would be required training for future controllers.
A review of training records for controllers involved in the accident did not indicate that they received TEM training in 2017.
Nor were controllers familiar with the term threat and error management during post accident interviews.
TEM training can strengthen situation awareness, promote team communication, emphasize effective scanning habits, help in recognizing patterns and developing adverse events, and enhance decision-making under stress.
TEM training would also benefit controllers performing supervisory duties who are responsible for overseeing facility operations and making operational decisions such as when to combine or decombine control positions.
Staff have proposed one recommendation to the FAA in this area.
When making operational decisions, supervisors must balance safety and risk management with the operational demands of the facility.
The DCA ATC tower standard operating procedures did not provide guidance or make a tool available to support supervisors in identifying risks, analyzing the impact of risks, prioritizing risks based on the likelihood and impact, or developing strategies to reduce or eliminate risks.
A real-time risk assessment or decision-making tool would have likely benefited the operation supervisor in identifying and mitigating the operational risk factors that were present on the night of the accident.
Staff have proposed one recommendation to the FAA in this area.
Thank you.
Dr.
Kayagel will now continue with his presentation.
>> Good morning.
I will discuss post-acc drug and alcohol testing requirements for FAA employed air traffic controllers.
I also will discuss the postac drug and alcohol testing of the local controller, assistant local controller and operation supervisor involved in this accident.
FAA employed air traffic controllers are subject to federal workplace drug and alcohol testing requirements for Department of Transportation employees.
These requirements are established by DOT order and are distinct from the drug and alcohol testing regulations that apply to DOT regulated employers such as airlines.
By DOT order, postac or post incident drug and alcohol testing is required after an accident or incident that involves any of the following: a fatality, substantial damage to aircraft, vehicles, or property, required medical treatment away from the accident site, or other unsafe practices as defined by orders or regulations.
Controllers are identified as being subject to required testing when their performance provides reason to believe that it may have contributed or when their performance cannot be completely discounted as a contributing factor.
By DOT order, required postac drug and alcohol testing must be conducted as soon as possible following an accident.
Whenever possible, required post-acc drug testing must be completed within four hours of the accident and required post-acc alcohol testing must be completed within 2 hours.
The DOT order allows for drug testing up to 5 days after the accident and alcohol testing up to 8 hours after the accident.
The controllers in this accident underwent workplace postac urine drug testing about 18 to 20 hours after the accident.
This testing did not find any evidence of prohibited substance use.
Postac alcohol testing although required was not performed.
This investigation determined that the FAA air traffic organization made the drug and alcohol testing determination almost 3 and 1/2 hours after the accident.
The decision was made to test for drugs only because the controllers already had been released from the facility about 15 minutes earlier.
Drug testing was scheduled for the afternoon of the day after the accident.
staff found that the ATO's drug and alcohol testing determination was not sufficiently timely to meet DO's required testing time frames.
Additionally, the ATO's decision to not conduct drug testing as soon as possible and to not conduct alcohol testing at all violated DOT requirements.
The ATO's delayed and inappropriate drug and alcohol testing determination was due in part to adequate inadequate ATO staff understanding of the DO's requirements for timely testing.
Staff also found that the ATO's initial event response procedures are inadequately designed to meet requirements for timely testing because the ATO procedures dictate that the drug and alcohol testing determination be made concurrently with the services rendered telephone conference.
The SRT is a management review to assess air traffic services associated with an event and does not take place until after multiple time-consuming administrative and initial investigative actions, including notifying and convening various personnel and preparing audio communications and radar display information for playback.
Staff have proposed two recommendations to the FAA and one to the DOT in this area.
This concludes my presentation and we are now ready to answer your questions.
>> Thank you all very much.
Thank you Mr.
Banning and certainly um Mr.
Soer, Dr.
Wilson and Dr.
Kaigel for your excellent and thorough presentations.
Um as I mentioned at the outset there are a lot there's a lot of information in the report uh substantially more information than we're able to present in even a day's uh time.
Uh so uh we have had the pleasure of uh re reviewing your work uh your in the draft and uh appreciate that.
I have a question.
I I have a lot of questions in this area but one I want to start with.
Ronald Reagan Washington National Airport opened in 1941.
Runway 1533 was part of the airport's original configuration since it opened.
Uh the Baltimore Washington helicopter route chart was published in 1986 maybe went through one small change to add a compulsory reporting point around Wilson Bridge.
Uh between 1991 and the timing of the accident there were there had been changes to route altitudes and police zone boundaries but not any substantial changes to the route.
You mentioned that the routes were never designed to ensure separation in your presentation.
Um yet we know over time concerns were raised repeatedly went unheard, squashed, however you want to put it, stuck in red tape and bureaucracy of a very large organization somehow didn't make it up.
Repeated recommendations over the years.
So much so that DC Tower took it upon themselves to create a helicopter working group because they had concerns.
Uh, how is it that no one, absolutely no one in the FAA did the work to figure out there was only 75 ft at best 75 ft of vertical separation between a helicopter on Route 4 and an airplane landing on runway 33? How is it nobody until we did this investigation? >> That's a great question and we've asked the same question amongst staff and that's precisely why we have recommended the assessment of these helicopter routes.
Not just what was there at the time but everything that exists today uh for this airport.
They've conducted others but um we're focused on this one right here.
So we have recommended that and our recommendations.
>> Yeah.
In fact, in our in our findings of our urgent recommendation report, we started ex we stated existing separation distances between helicopter traffic operating on route 4 and aircraft landing on runway 33 are insufficient and pose an intolerable risk to aviation safety by increasing the chances of a mid-air collision.
I'm going to ask Mr.
Banning and uh Mr.
Learon, your pilots.
Is there anywhere in our airspace that 75 ft of vertical separation is acceptable? >> I'd say not to me as a pilot.
>> Yeah.
Not that I'm aware of.
>> Absolutely not.
>> Absolutely not.
But it wasn't until uh Miss Mohler and others mapped it out.
Thank you for your work.
That we figured it out.
Get years of nobody listening.
Opportunities to.
So I'm going to ask FAA order uh states that terminal operations service area directors are responsible for conducting annual reviews of existing helicopter route charts to determine their accuracy and continued utility.
Uh what uh do we have any evidence that these annual reviews were conducted? >> No, we do not.
We asked for that information and they were unable to provide that or nor identify the individual that you just named.
>> Right.
>> Since then it has been changed to the air traffic manager responsible for those.
But at the time of this event it was that individual that you named and we have no evidence of any of it.
>> Do we even know who the terminal So we have no idea who the terminal operations service area director is for the Baltimore Washington helicopter route chart.
That was the requirement and the person responsible for conducting these annual reviews at that time.
>> That's correct.
We do not know and as of two days ago, the FAA responded in the same manner.
They did not either.
The FAA had no idea who a terminal uh what is this terminal operation service area director was.
Who was responsible for conducting annual reviews ensuring the safety of helicopters transiting the airspace around DCA? No idea who that was.
Yeah.
So, we invited an individual to our hearing uh from the mapping office to our investigative hearing to discuss the route to discuss annual re reviews and discuss any sort of proposal.
Turns out we learned after the hearing she wasn't the right person to invite.
We should have invited this terminal operation service area director that nobody at FAA could figure out who that was.
Uh was FAA a party to this hearing? Yes, they were.
>> What is required of parties to the hearing when we come to an hearing? For example, are they part of the planning of the hearing? And would they go through the witnesses and list of witnesses with us and possibly be required to identify when we have the wrong witness? >> Absolutely.
They are to they their responsibility is to advise us on the the correct witness for the correct uh subject area.
Did FAA raise this with us at all that this was not the right person to question at the time? >> They did not.
They actually provided this person to us.
>> They provided this person to us that could not answer any of our questions at the hearing regarding annual reviews probably because they had no idea who a terminal operation service area director was.
Thank you.
Uh that runs out my five minutes.
So, uh, turning to member, um, I am so sorry, Mike Graham.
>> It's been a long few weeks, hasn't it? Uh, I'd like to, uh, direct these questions to, uh, Mr.
Soer and I'm going to ask you some questions about the primary duties and responsibilities for the different control positions in the control tower as they relate to the DCA ATC uh tower uh information outlined in their SOP.
As a former controller yourself, I want to I want you to share your perspective on which duty or responsibility is most important for each of the following positions.
What would that be for the local controller? >> According to the DCA's SOP, air traffic control towers SOP, the local controller must provide initial separation between su between success of departures and arrivals and provide separation between arrivals and departures and those aircraft operating under their jurisdiction airspace.
That's their primary duty.
>> Yes.
How about the helicopter controller? The helicopter control according to the SOP is to separate VFR traffic from the DCA arrivals and departures or fixedwing traffic, issue safety alerts and traffic advisories as required, and to clear VFR aircraft on routes or into zones as depicted on the Baltimore Washington helicopter route chart.
>> How about the assistant local controller? They are there exactly like you might think to alert the local controller or assist and alert the local controller of any unusual situations or traffic developing traffic conflicts.
Maintain surveillance of the local traffic pattern and landing areas and utilize uh visual memory aids when required to help them not miss anything and assist the local control and monitoring aircraft uh in their airspace as well as on the CTRD or the control tower radar display.
And how about the primary duty and responsibility for the operation supervisor? The operation supervisor provides operational supervision and directs the tower operation to ensure an efficient flow of air traffic, making necessary notifications of suspected operational errors, pilot deviations, near midair collision reports, other incidents requiring notifications, combining and decombining of a of positions in accordance with criteria outlined in their SOP.
>> Thank you.
Can you share a brief summary of how the decision was supposed to be made according to the SOP in effect at the time of the accident to combine or decompine the local and helicopter control positions? Yes.
In the SOP, it outlined uh that the supervisor or controller in charge at their discretion um though it stated, let me start out first with it stated that the helicopter position must normally be decombined.
So, normally it should be decombined during these core hours period.
If the supervisor or CIC has a need to combine or decombine, if they were already combined, the positions there to consider.
They have a list of seven things which I can list here for you.
the weather conditions, VIP movements, special helicopter operations, training initiatives, staffing constraints, air carrier traffic volume, helicopter traffic volume with the within the DCA class Bravo.
So when making a decision to combine or decombine, they should be taking all of these into account.
>> And who makes that decision to combine or decombine? >> At that time, it was the operations supervisor.
>> Okay.
uh if they were to decombine it, which they were not at the time of the accident, did they have enough personnel on the tower to do it? They did.
They did.
Okay.
Um so based on the the procedures and the present staffing levels in the tower with 12 aircraft total on frequency in the 90 seconds leading up to the accident as well as a steady stream of helicopter traffic in the airspace all evening.
In your op opinion should the local and helicopter control positions had been decombined at the time of the accident? staff looked at this very closely and we agree that uh the we believe they should have been decombined in accordance with their manual however softly it may have been written and that we focused some recommendation was actually in this area because we're concerned about that combining of positions right one last question I know there's a good discussion of this in the in the draft report of the operation supervisor's extended time on position without a break and I I think Dr.
Wilson had talked about this.
Um, based on your investigation, are there any other factors you see as potentially playing a role in the operation super supervisor's loss of situational awareness and the reduction in in their ability to proactively assess the risk posed by the traffic and environmental conditions at the time of the accident.
We looked at um a number of factors that could have impacted his performance, but we or his awareness um on duty that day, but we really felt that his extended time on position uh was the reason that he had the decreased vigilance and attentiveness.
>> And was there anything looked at as the amount of time that they had been at that facility? Could you clarify what you mean by >> that supervisor? How long have they had been at that facility and making this their tenure there? >> Uh >> that's okay.
I'm past my time.
>> I I understand what you're saying.
So yeah, he he was a relatively new operations supervisor.
Um and one of the recommendations that we have particularly for the threat and air management is we think that that would have benefited him on uh making better decisions that day as well.
>> But they had prior experience as a controller.
Correct.
>> Yes, he did.
He was new to being an operation supervisor but had been a controller for some time.
>> Thank you.
>> Member Enman >> Kad I want to follow up just a little bit on that because you mentioned threat and error management.
We made this recommendation in 2016 based on something we had seen and the response we got was yes, we're going to do it and all new controllers are going to get it.
We were promised that this was happening.
Right.
>> We were.
>> Did they do it? >> When we reviewed the training records for the controllers involved, we did not see that they received the integrated safety training workshop which included threat and error management training.
Uh we did see that the operation supervisor did have a workshop of that same name several years um a couple of years before and then a new hireer who was working on position not one of the controllers involved.
He had a workshop in I believe 2023 with that same name but that was the only evidence that we had of that workshop being provided to any of the controllers on duty.
>> So you just mentioned two people.
>> Yes.
And there were 21 that were assigned to the tower.
>> We only looked at the training records for the controllers that were involved the night of the accident.
>> Okay.
But obviously what we what we were told that has happened and was occurring did not actually occur.
>> We don't have evidence that it occurred.
>> It's okay.
I can say that.
Thanks, Cat.
>> The answer is no.
>> It's kind of like congressional testimony.
I'll get you there eventually.
Dr.
Dr.
Killil, uh, you mentioned about alcohol and drug testing, and I know we don't know or don't believe that it may have played a role in this, but it's obviously a violation and something we've looked at many times before.
Is that correct? >> Uh, this is not the first time that we've seen it.
And, um, you're correct.
We we did not find any evidence that this was an issue in this accident, but we did address it as a safety issue in our recommendations.
>> So, you saw the investigative hearing, correct? >> Yes, sir.
And Nick Fuller testified at that point that they were aware of their shortcomings and they were going to make modifications and changes.
Is that right? >> That's correct.
Mr.
Fuller testified that the FAA was working to revise the uh FAA's order for initial event response procedures, which are the orders that um tie the drug and alcohol determination to the services rendered telephone conference.
>> And do we believe do we know if they've done that yet? >> As of this date, they have not published that as a revised procedure.
No, sir.
>> Exactly.
But the answer to the question is no.
because I was in Louisville, Kentucky when 15 people died and they didn't test then either.
So, they're not learning a lesson obviously, nor what they're telling us seems to be factual all the time.
It's okay.
It's a rhetorical question.
Dr.
Gail, um I want to go back, I guess, Mr.
Soer, uh I I asked or question this about putting this into the report, and I understand staff's reservation because it's considered slang, but can you tell us what squeeze play is? It was mentioned in several of controller interviews and it was discussed somewhat in the reports.
>> Yes.
So there is no um definition formal or informal for squeeze play.
It's a it is sort of a slang term but controllers understand it for those times when you are um taking action to get a departure out in between arrivals.
that might be a tight space or to fit one into an arrival sequence and it may be in a radar situation where they're having to expedite everything and everything has to kind of work just right for it to >> So you I think you're describing this type of scenario that happened with 5342.
Is that correct? >> Um 5342.
Yeah, somewhat.
Yeah, it's it's similar to a squeeze play.
Yes.
>> So I'm kind of curious.
Um the controller had asked them to circle so they could offload a departure.
But we look back and American 1630 was the last aircraft to be cleared for takeoff prior to the accident and it departed 12 minutes early.
They have safely arrived 25 minutes early.
I guess do the controllers know when they're prioritizing an aircraft so it can leave earlier or should they even be aware whenever they're trying to attempt this or are they trying to just I know we've heard some testimony in regard to the geography at DCA but should they be aware of that? >> So they would generally not necessarily be aware if they were early.
They would be aware if they had been assigned a departure time based on traffic initiatives or something where they actually get an assigned time.
So in those cases, they would know they they have a they have to get this aircraft out in a particular window.
As far as if a aircraft had taxied maybe a little bit early, they might see the fact that he was scheduled for departure at 2020 and he was ready for departure at 2015 or something.
They they might see that, but it wouldn't necessarily alarm them or I mean change something.
>> Okay, I'll wait for the next round.
>> Thank you.
at the time the helicopter and local control um positions were combined.
Uh are uh are the communications for air traffic control all on speaker >> for local control? Yes, they are.
At this tower, they keep local control communications on speaker all of the time.
>> So the operations supervisor could hear what was being stated.
He should have been able to.
>> Okay.
Um, PTOIC, my understanding is that at some point the operations supervisor received a phone call from PTOIC Tron uh requesting a reduction in miles and trail.
Correct.
>> That is correct.
>> And essentially what that means is to reduce spacing to allow for more arrivals.
>> Correct.
What does that do uh for uh uh for the controllers? >> When you start decreasing below that u miles and trail that was requested or that was in for the restriction, it makes it much more difficult to get departures out in between those arrivals.
>> And when they accept that, is there any sort of criteria or look or did the operation supervisor say to the local controller, hey, you know, is this okay? How do you feel about this? or it was just approved.
>> We have no evidence that he asked anybody and according to the recording, it was relatively quick.
When he asked, he responded immediately that it was approved.
>> That it was approved.
So, they would reduce spacing and get more arrivals in increasing workload for the controllers.
What was the assistant local controller doing at this time? at the time that they made the decision to >> I'm sorry at the time of the accident time of the collision.
>> So at the time of the accident according to post accident interviews the assistant lo controller was writing down information on helicopters >> and what is the role of the assistant local controller >> to assist the lobe controller in surveilling the airborne aircraft and monitoring the CTRD for conflicts and alerts.
>> Monitoring and helping alert.
>> Correct.
>> Okay.
Thank you.
Uh, how was the mix of it? Actually, I'm going to read this from our report because I don't want to catch uh offguard because you don't have the full report right in front of you.
Um, but I do think it's worth noting that in our report we state DCA is capacity capacity constrained.
Since 1991, annual air carrier tower operations increased from 175,224 to a peak of 294,312 in 2024.
Over the same period, tower operations attributable to air taxi and general aviation declined substantially.
Air taxi operations peaked in 2007 at 118,228 operations, but fell to 1,594 operations in 2024.
General aviation operations declined sharply after 2001 from 32,286 to fewer than 3,000 operations per year in every subsequent year.
Air taxi GA meaning general aviation and military operations accounted for 43.8% of all fixedwing uh instrument flight rules operations in 2007 but only 1.3% by 2025.
This change also off altered the mix of traffic operating at DCA with smaller airplanes increasingly replaced by larger air carrier category aircraft because larger heavier airplanes have more restrictive performance margins on DCA's shorter runways.
Demand for runways 119 the airport's longest runway increased.
What did this change in the mix of traffic do uh to did this increase workload for the tower uh and any sort of co coordination demands for local and ground controllers? >> It certainly did.
Uh we've documented in the report that the uh as we mentioned earlier even in my presentation they have the busiest single runway in the national airspace system and what you just explained is is precisely why >> and they were using runway 33 to offload some of that traffic in order to get more in and to deal with some spacing including other measures like tromboning and other things which we could talk about later.
But my point is this did increase workload and did anyone in FAA ever say, "Hey, maybe we should look about look at how things have changed at DCA." >> We have found no evidence of that and have made d recommendations directly to that.
>> Yeah.
Well, I wonder if they actually looked at any other airspace or major airports in the US.
With that, uh, member Graham, >> thank you, Chair.
I think these are for you, Mr.
Soer.
Um, how often are airport arrival rates reviewed? According to uh FAA procedure, they should be reviewed annually every February.
>> Okay.
And what factors go into those reviews to determine if an arrival rate needs to be reduced? >> I'm going to look at my notes real quick.
>> No problem.
>> So, there is a list and that's why I knew I wanted to look at my notes.
I had this list handy because I thought this may come up.
So they look at a a list of things that they have listed.
Intersecting arrival departure runways, distance between arrival runways, dualpurpose runways um for when they share arrivals and departures to the same runway, land and hold short utilization, availability of high-speed taxiways, airspace limitations and constraints, procedural limitations for missed approach protection, noise abatement, etc.
taxiway layouts and meteorological conditions also be considered >> and all those factors in context with the DCA.
Would you say that every one of those factors was in play with the arrival rate at DCA? >> I think our staff concluded that it does not appear that and we've have focused recommendation for that very thing.
>> Okay.
>> Because we think it needs to be reviewed.
>> Can I ask for uh table 18 to be pulled up? And as I do that, I know it's going to take a few minutes.
I got a couple other questions as we before that.
Uh, what was DCA's arrival rate at the time of the accident in VFR conditions? >> 36 at the time of the accident.
>> Okay.
Did the FAA lower DCA's arrival rate in the years leading up to this accident? >> There had been temporary lowerings of their arrival rate, but nothing's uh permanent.
Were there requests made to the FAA to lower lower the arrival rate at DCA prior to this accident? >> Yes.
>> Okay.
Can you explain in detail the request that was made to the FAA to lower lower the arrival rate and when that was made and who made it and why did they make it? >> Yes.
In May of 2023, actually, as we heard in our investigative hearing and and we spoke to the this individual as well as part of our post accident uh investigations, the air traffic manager from PTOIC Tron um submitted a request to lower the arrival rate uh specifically due to um the fleet mix change that had occurred over the years and the fact that it didn't appear the arrival rate had been reviewed in quite some time as well as the challenges they were having with miles and trail and being able to provide adequate miles and trail um and the limited constraints they have on the surface area there.
It was getting difficult to be able to keep up.
So they submitted a request to lower the arrival rate at that time.
>> And how did the FAA respond? >> We were unable to locate the final but uh ultimately they did nothing in response to that.
>> What's an arrival rate today after the accident? >> It is 30 today.
>> Okay.
So they lowered it now, but they didn't in 2023 after the request.
>> Correct.
And it has fluctuated since the accident.
>> You got table 18.
Please put that up there.
You can see there.
Uh this is the uh percentage of flight offloaded per year to runway 33 when landing in the northbound configuration.
Is that correct? I'm not sure.
It's not that they necessarily the numbers that were offloaded, but it is the percent of usage of runway 33 in total operations.
>> You see it's used a little less now.
Can you explain why? >> Absolutely.
I mean that's that's the fleet mix that we keep talking about.
Different type aircraft, less aircraft can take that runway because of its shorter distance.
>> Okay.
Um, I know you had a uh are there any other ways that a controller can offload uh going to runway one other than runway 33? >> Sure, there are mitigations that a controller can apply whether it's uh speed adjustments, uh some sequencing, they can even give turns.
I know you you the chair mentioned tromboning is which is something that occurs out in the radar airspace but turns in space or turns and speed adjustments can but they're they're not as effective.
So they they use 33 most often because it it works very well.
>> Okay.
And I know you had a very good presentation on the slot program and the front end loading and backend loading and all that.
My final question for this round here is uh if we tie this all together, what impact does a higher arrival rate than an airport can sustain, less alternatives to offload traffic due to larger aircraft and compressed airline scheduling have on air traffic controllers? It increases all of their workload and increases risk >> and it also makes them have to do squeeze plays and make it work.
a lot of mitigation mitigation activities.
>> Okay.
Thank you, >> Member Inman.
>> So, in that regard, we we heard some testimony earlier that uh the local controller stated they were a little overwhelmed about 15 minutes before the actual crash.
But then we also heard that 90 seconds before the volume had actually increased.
How does the controller get out of that situation? The controller could have asked to have the positions decombined.
>> And I guess from a practical point of view, you've got people that are in the tower working at that moment and then you got people that are in a break room that are several floors below.
Is that correct? >> Correct.
>> So if I'm 90 seconds, my volume picks up and I say I need help.
How long is it going to take to get somebody up there to help them? >> It could take minutes.
Um, but we could also he could also rely on the assistant local control and also the operations supervisor to act as extra set of eyes and to assist in any way that they could in terms of monitoring traffic until somebody could get up there for relief.
>> But the assistant local controller was already supposed to be the other set of eyes.
Right.
>> She was an extra set of eyes.
Yes.
Had she known that the controller was uh overloaded and needed help, she also could have um potentially stopped writing down helicopter information and also uh started monitoring traffic for him if he felt that there was an area that she needed to focus on.
>> And in this case, the uh the role of that supervisor is to be monitoring all of this.
Is that correct? >> Yes.
>> So, what role did they have in in monitoring this? They were on station for 4 hours and I know we kind of talked about that but in this case the supervisor should be the one who starts to realize it before anyone else.
Right.
>> Yes.
Which is why we've recommended the threat and error management training which we believe would help with that as well as a risk assessment tool that they can use in real time.
>> So is that supervisor one that actually received it out of the five? >> He received a workshop with that same name a year or two before.
I think it was 2014.
So, it was technically before the 2017 um response that we had from the FAA.
So, we don't know what was included in his 2014 training.
>> And so, it sounds like he's been around for a long time.
>> He was hired in 2011.
>> Okay.
I guess how does a how does someone monitor whether a supervisor is actually actively supervising there would be no one there I guess in real time monitoring the supervisor.
>> So there's no real way of knowing that these supervisors are actively engaging unless it's brought up in other reviews.
Is that correct? >> Correct.
>> Okay.
Um I'm going to go to just and this is for a reference for later.
Uh and I've got a reference and that we can pull up and I don't know if we need it.
I think Brian, you know, 14 CFR 91.123.
Uh, whenever it talks about ATC clearance obtained.
Great.
Thank you.
Um, it says no pilot in command may deviate from that clearance unless an amended clearance is obtained.
An emergency exist or the deviation is in response to a traffic alert and collision avoidance system resolution advisory.
So, I guess my question is in that regard, we're going to talk about TCAST and uh ADSB later, but in this instance, um if they had gotten some type of warning, the pilots, could they have made any change or does it have to be what ATC actually tells them? >> No, the pilot is always in final command of the aircraft.
And in that case, I think we're talking about a traffic advis or a uh a u collision avoidance uh alert to them with resolution advisory.
And in that case, it meets that exactly where they will respond to that and they can break away from that clearance.
>> Okay.
And we may get to that in the next section.
Do we know what the ATC scope showed at the time that this incident occurred? Because I know we we've had discussions about whether it rounds up and rounds down.
>> Yes.
air traffic altitudes displayed on radar are rounded to the nearest 100 ft.
And at the time of the the closest time to impact that we can see there, the helicopter was shown at 300 and the CRJ was shown at 200.
>> Wait, I got that wrong.
Hold on.
>> Can I let me check that? >> They were both three and four.
We we can we can look and check on that, but essentially we believe that the uh the CRJ showed above the helicopter, but the helicopter climbed right at the right kind of right just before impact.
It showed a climb from 200 to 300.
I do know that.
>> Okay.
So, you'll just we can go on the next round.
You can maybe answer that for me if that's okay.
>> Yes.
Okay, great.
Um, I want to, um, well, one, I've been asked a lot about staffing.
Obviously, we felt that the tower, you determined the tower was adequately staffed, though certainly the workload was significant given positions were combined.
But I do want to provide I asked the staff to check based on uh September 29th, 2025 versus you know today.
I think we looked at January 5th, 2026 and what those uh numbers were.
And I'm just going to go for, you know, certainly there are more authorized positions, but I think I'm going to focus on who was operational, meaning numbers of people available to work that are working in the tower, not what's authorized, but that are working on the tower.
And so, you know, by my the um uh the draft report shows for certified professional controllers on January 29th, operational was 2025.
I mean sorry 25 today 20 traffic management coordinators they had two today zero operational operations supervisors four on that date today or as of January 5th three operational operations managers one on uh January 29th 2025 uh this just a few weeks ago two support manager one at that Time now to air traffic manager still at one temporary was what existed on January 29th, 2025 today.
Still one temporary staff support specialists.
Uh although authorized to have three, there are zero.
Uh so by my calculations, you actually have a lower number of personnel today that are operational in the tower than existed on January 29th.
It's not a question.
It's just uh we provide that in in the uh report and so we're not going to talk about it but I want to make sure uh that folks are aware of that.
Certainly there are several trainees in the process but we are at a lower number than we are today that are operational.
Um with respect to the facility level I do want to get into that.
Um but talk about a a little bit on it.
It was a level 10 facility.
Now it's a level 9.
Uh why was it changed? >> It was due to a change in the way they computed traffic counts.
>> So they had traffic counts for PTOIC Tron.
They had p traffic counts for the tower um which included helicopters.
And uh at some point they said, "You know what? We're going to give all those counts to PTOIC Tron.
We're going to take it away from DC Tower." Once they took those counts away, the level decreased to a nine.
>> That is correct.
And PTOICS increased to a 12.
>> And PTOICS increased to a 12.
Here's my question.
Is there a difference between the traffic that PTOIC Trcon handles and the amount of traffic DCA Tower handles or is it the same traffic? >> They're still operating the same traffic in the same manner they were prior to the change.
>> Exactly.
So, this change was made.
They downgrade the facility.
Obviously, you're talking about things like cost of living and other things like that.
Uh but from a safety standpoint, how does that de decreasing a facility a tower level in in uh decrease safety? How does it impact safety? >> As we mentioned in the report, the the the biggest thing that it that happens there is they cannot attract the experience or get the talent into the facility that is need to run a very complex air traffic control operation.
This is the most complex airspace, one of the most in the entire United States.
And the and I'm not I'm not uh uh minimizing the experience of tower personnel uh at DCA Tower, but if you're going to attract personnel and have the most highly experienced personnel, you want a higher level facility.
Why? because they're paid by facility level.
So the higher facility number, the more pay you get.
In this case, everyone in the tower knows this tower is essentially treated as a training facility.
You come in, you do a little bit of time, then you go out after you get your time because you want to go get paid at a level 10, 11, 12 facility.
So it is a pass through.
they can't attract people here.
Uh, which is, uh, a problem and which is why we're making some recommendations on that.
Member Graham.
Thank you, Chair.
Uh, Mr.
So, is there a minimum distance between converging aircraft where a controller can approve a request for pilot applied visual separation or do controllers have the latitude to approve requests for the pilot applied visual separation whenever they deem appropriate? So when a when a controller is going to approve the application of visual separation, they must have approved required separ standard separation prior to that.
So in the case of DCA, they're a class Bravo airspace between a helicopter and a VFR helicopter and IFR fixed wing.
they would be required to have one and a half miles of separation or 500 feet vertical separation and before up up until the point they approve visual separation.
>> Okay.
And um so after the pilot applied visual separation is approved in class B airspace.
You you said it's uh the minimum steps a mile and a half and 500 ft.
Correct.
>> Not not after it's prior to prior to approving it.
Once it's approved, then it becomes the pilot's responsibility for that distance.
>> Is there a distance in there that they if they decrease that range that that's something the controller has to do at that point? >> No.
>> Okay.
So, when would a controller make a safety alert? >> Anytime the controller observes an aircraft that they believe to be in unsafe proximity to terrain or another aircraft.
And is that a is there a design there's not a designated distance or vertical separation for that? Is that correct? >> That is correct.
>> That's correct.
Okay.
What does a safety alert entail? >> Safety alert would include the the words traffic in a case of between two aircraft.
It would be the words traffic alert followed by an appropriate traffic advisory followed by a alternate course of action if feasible and then the word immediately.
>> Should these two aircraft in this accident have received a safety alert? >> Yes, we've stated that in ours that the in the report that the controller should have issued a safe would have been the most appropriate thing at that time.
>> Okay.
Um, switch here a little bit.
Um, there was a lot of impacts on the higher arrival rate coming into DCA, changing fleet mix.
We talked about compressed airline scheduling practices, uh, the controllers and their ability or inability to continue following SOPs in order to make it work.
Would you agree that one of the consequences of the cumulative effect was the deviations in traffic calls, phraseology, application of visual separation, call signs, radiocom, training records, the list goes on, and other issues that had become standard and were identified in multiple external compliance verifications of DCA air traffic control tower.
I think I understand your question and I would say that uh we've we've outlined in the report and staff believes that the continually increased workload uh for many reasons whether you're talking about from the traffic to the combining of positions inappropriately to whatever at the time um we believe that led to uh the ability for a lot of deficiencies to occur or operating as we heard in the during the investigative hearing, the operations manager said specifically, "We're operating on the on that borderline of safety regularly." So, we made recommendations obviously focused in that area.
>> Okay.
So, um do you think the controllers at DCA had fallen into a state of relying on pilot applied visual separation as a crutch despite its many well-known and well doumented shortcomings? staff does believe that there isn't there was an over reliance on the use of visual separation as a means of providing separation for aircraft and moving aircraft through the airspace so regularly.
>> Okay.
>> Yes.
Thank you.
And finally, um the last communication between the controller and Pat25 um was was that an appropriate safety alert? It was not a safety alert by definition.
Uh but it was an attempt at the time to deconlict.
>> Thank you.
>> Thank you, member Enman.
>> You have an answer for me, Ryan, or you haven't maybe had a chance.
Sorry.
I do have an answer for you.
According to the stars maintenance replay, uh the um PSA flight 5342 was at 400 and PAT25 was at 200.
The moment targets merged, we had 400 and 300.
And so the controller, if they were looking at their scope, would have seen that at least indicated on the scope that it was 100 ft or approximately 100 feet depending on the averaging of uh the recommended route maximum altitude.
Well, it had actually been at 200 ft for a brief time there where he was at 200 right up until he he was at 300 then 200 for a little bit and then came up to three right when targets merged.
>> Okay.
So, and that would have uh made a conflict alert.
Correct.
>> The conflict alert was active at that time.
Yes.
So I think in your research uh I believe in the 30 minutes prior to this accident we have 18 controller radio transmission where a conflict alert is going off.
Does that sound accurate? >> It is accurate and that actually was appears to be five separate interactions that were occurring.
>> We had well not for the full 30 minutes.
We were only able to get replay information for 18 minutes prior, but in the 18 minutes leading up to it, >> there were five separate instances of conflict alert that uh resulted in nine times there was a trans there was a transmission with it in the background.
You could hear it.
>> So in that 18 minutes, there were that many times and that would have been heard in the entire tower cab.
>> Yes.
So, I go back to the supervisor.
If they're hearing a number of conflict alerts continually happening, should they not take action to say, is this a trend or is this just become a nuisance and they just avoid it? >> They should.
And I think that's exactly why as Cap as I'm sorry, Dr.
Wilson has mentioned previously, we uh we specifically have a wreck in this area for identifi for there's a potential for complacency to happen or for nuisance alerts.
And to try to fix that, we hope to we have a recommendation to actually adjust the conflict alert system logic.
>> Um we actually took a uh we we visited the tower last month.
Um, I know member Graham and I did and tried to see it in almost the exact same moonlight that was occurring on the night of the accident.
Can I take you back to that? And I the question I have and it was on the uh the simulation that was being done >> and the pictures indicated these monitors that were hanging from the ceiling.
And if you look very closely as you look at the converging paths, it's almost that the monitors are blocking the ability to see depending on how high or short or tall you are.
Is that a problem in just a tower configuration issue? >> We looked at that and actually I'm going to let Dr.
Wilson speak to that because she has some information in that area.
>> I think it's just important to note that those monitors are movable so they can be adjusted.
And also when we talked to the controllers who were on duty, everyone said that they had no issue um identifying both aircraft that there were no um limitations in them being able to see.
So they didn't feel that the monitors were in a position that blocked their view from monitoring both aircraft.
>> Okay.
And and during that tour um Brian did we we talked about whether the the morale and the staffing had increased or at least the perception of the training uh for that tower.
Correct.
>> Yes.
>> During that discussion, did we hear anything about any other towers that might be having the same issues that DCA was having at the time of the crash? I don't recall specific ones, but I do remember uh maybe one or two of the people from the FAA just kind of uh socializing that there's there's several, you know, there's facilities across the NASA experiencing staffing problems and training deficiencies and the ability to to staff appropriately.
>> Yeah, I think that's my concern is we were talking about they may be getting the time and attention and the training, but is there other places in the United States? Are we going to sit here 5 years from now and say the data was there 5 years ago because we know now the data the people were raising the concerns.
People were saying this was dangerous 5 10 years ago and nobody was really listening.
My concern is is this hap is this somewhere else in the country and do we know that they're getting the proper resources? Our hope uh staff's hope is the recommendations that we have in these specific areas will not just be focused on DCA and the air traffic organization will organically look at that and say we need to look everywhere because it does apply.
>> How's their track record been in the past? Come on, Brian.
This is your chance.
>> Not great.
>> Okay.
Thank you.
>> All right.
Thank you very much.
Um, I actually want to ask um along the lines of where member Enman was because I I do have this question about the operations supervisor.
It was on speaker all the conversations.
I mean all the communications going back and forth.
He took a call to then uh reduce essentially miles and trail and approve that reduction by PTOIC Tron.
And I I want to point out, you know, something in our report.
In the 20 minutes before the accident, the total number of aircraft the local controller was handling fluctuated between 7 and 12.
At about 10 to 15 minutes beforehand, he said he was feeling a little overwhelmed.
The operations supervisor said it was totally fine, no problem.
I uh I was a little bit surprised by that.
I mean even if you have procedures in place on decombining you've got how do you even capture to one person he's fine to that local controller who's busy busy busy and maybe doesn't have two seconds to say hey let's deconlict while the operation supervisor is listening to all of this everybody in the tower is hearing it uh I mean how is that not complacency I mean we talk about reduced alertness and vigilance, but I raised this earlier about is it complacency.
You just mentioned complacency actually, Mr.
Sober in your answer to member.
>> I do I do agree.
I do believe and I mean I think that staff believes there was complacency occurring in that tower on some level.
But again, not to keep going back to them, but that's actually the primary reason for our focused recommendation in the area of giving them the tools that they should need, the abil the requirement to log when they combine positions.
So, there's some sort of accountability and tools available to the supervisor to be able to make better decisions and understand what their true responsibilities are while they're there instead of just kind of here you go, you got it.
he had received very little supervisory training as we from what was documented from the FAA uh to prepare him for that job and so we made recommendations there we hope would improve that.
>> Well I think when we get to those findings and recommendations I may have some more questions on that.
Um M you talked about uh member Graham had asked about um what are essentially merging target procedures where you do issue a safety alert uh at what point and member Graham had asked about um a communication you know at first it was do you have do you you know do you do you have visual on that do you see do you have visual on that CRJ first of all that you can't distinguish between a CRJ700 and anything else in the dark with night vision goggles at that point.
Uh but at at uh 8:47 and 41.9 seconds uh the communication of pass behind the CRJ was made.
>> Yes.
>> Okay.
So 8:47 then the collision is 84759.
So we've got about 18 seconds in between.
Is that when that uh safety alert should that safety alert been issued instead of pass behind the C CRJ? Is that when I'm asking when that should have been issued? >> So from from the stuff there, the time in which the safety alert should have been issued was when he first asked him, do you still have that CRJ in sight? >> Okay.
So uh 8:47 39 seconds.
So that does provides about 20 seconds.
>> Yes.
Okay.
Um, my last question is, given that no one, apparently no one we interviewed was aware of routine close encounters of helicopters other than people in the tower, was any organization, any organization in your opinion, effectively monitoring the potential for the risk of a mid-air collision between military h helicopters and civilian aircraft in that area? I believe that's going to be discussed with some depth in the in the fi in the third panel today.
>> In the third panel, we'll go with that back in the third panel.
And Dr.
Price, thank you for that.
Um, okay.
Do you have any any additional questions? >> You have one more member Graham, >> let's talk about the the supervisor at the time.
This is was brought up about the situational awareness.
What did what did they say in the interview uh when asked about how many helicopters they thought were on on frequency at the time? >> Oh, I'm sorry.
Yeah, the supervisor only recalled one helicopter, >> but yeah, leading up to it, there were >> five.
>> It varies between Yeah, that's >> considerably different.
>> Yeah, the loss of situational awareness, I'd say.
Thank you.
That's all I have.
Member Emman, >> I've got just a little bit of followup and I want to go back to the helicopter route structure itself and and this was a little bit baffling to me that we got so many different answers from the FAA.
>> Is there a maximum altitude for a helicopter route? >> Yes, they do have maximum altitudes published on the chart as we found.
>> Then why did their testimony say it was only recommended? because they had conflicting information on the helicopter route chart as we outlined in the report and produce recommendations to hopefully correct that rec that in one place it said altitudes are recommended and then in another place it said altitudes are maximum >> where I'm from that's called talking out of both sides of your mouth >> right and and I mean just to add on to that the the letters of agreement with all of the helicopter operators are clear that the altitude ude maximum is what they are being cleared at.
They are to maintain that altitude through the airspace.
That's their that's their class bravo clearance actually.
>> Okay.
And the same thing in this route structure.
I I continue to be amazed as we learn more about it.
How can the Coast Guard tell me the exact depth of every piece of terrain that's on the PTOIC and yet someone tells us that this route is drawn with a Sharpie on one side and a riverbank on the other.
Again, we do have uh recommendations to that effect too because of the vagueness of that routing or to at least explain better so operators understand fully what their their responsibilities and restrictions are.
>> And I guess I'll go to Bryce if you don't mind just because being a pilot uh not going to let you off that easy.
Is it not concerning that I mean you you've have 16,000 hours I think >> about 8,000.
>> Sorry, double doubled it.
It's a long time ago.
8,000 hours.
Uh, is everything usually very precision in regard to not only landing approach plates, everything is lined out with GPS exactly where you have to be when you're on the bug.
Uh, IFR is that not just a science that's defined.
>> Yeah.
In the instrument flying word world world, it is very regulated and yes and it was disturbing to see that these helicopter routes had no lateral boundaries and the confusion on the charts with regard to altitudes.
>> So it was set up for failure just by the administration of what they were telling aviators.
it it was most certainly not clear and there was a misunderstanding about what those routes also provided which we'll I think discuss in more detail in another panel.
>> That's all I got for this one.
>> Thank you.
Just a a few for me.
Dr.
Kaill, um why is it important? You know, obviously we were not concerned about uh anybody having been impaired by alcohol.
We didn't think that was the case here, but you know, time and time again, they're not doing testing.
It wasn't just uh the UPS crash in Kentucky, but uh it was Austin, Texas, any number of uh investigations that we've conducted where time and again they're not doing testing.
Why is it important to do testing? Well, doing testing provides us with investigative information obviously, but beyond any single investigation, uh, a primary intent of drug and alcohol testing is to identify and deter, uh, use of certain prohibitive substances as well as abuse of alcohol with the recognition that that can impair performance of safety and security sensitive duties.
So anytime you have a systemic barrier to this getting done on a regular basis, routine basis as required, that's a safety issue.
>> It's a significant safety issue and it's one that could be left unchecked if they are not doing the required testing.
>> That's correct.
And that's why we have proposed several recommendations to try and um close this loop and make sure that this testing that's required is actually being done >> at the hearing.
Um, uh, Mr.
Fuller had stated, "Well, they didn't do testing cuz no one was in the tower." Mr.
Soer, how long after the collision was somebody in the tower from FAA? >> Within minutes? >> 18 minutes.
>> Yes.
>> Specifically.
>> Yes.
>> He also stated uh that uh they weren't quite sure there were any fatalities.
and that drug and alcohol testing was required when they can confirm fatalities.
We know that's not the only Was it clear there was substantial damage? Was anyone unclear that there was substantial damage? Anyone? >> I think um beyond what's obvious in this case, the the important thing to point out as you say is fatalities were not necessary for testing to be required.
That's >> right.
That's correct.
Um I have one last question um for Mr.
Soer or um or Mr.
Banning, who whoever wants to respond to this.
We have significant traffic changes over time.
We have no one who reviewed the routes.
We have routes designed that were not designed to ensure separation.
No safety reviews.
We have arrival rates increasing.
We have reductions in miles and trail including uh reductions in miles in trail that night which reduced spacing.
We have less and less alternatives where controllers are using other measures to deal with getting increasing uh planes in uh arrivals in due to pressure offloading onto runway 33, reducing speeds.
Uh we have increased workload.
We have a supervisor.
We have all of this that was on uh speaker.
We have a supervisor who didn't take action and then improved less spacing, more arrivals.
We have an assistant local controller who was who who had diverted attention to writing down something on helicopters at the time.
Uh, and we have an entire tower who took it upon themselves to try to raise concerns over and over and over and over again only to get squashed by management and everybody above them within FAA.
Were they set up for failure? >> Panel 3 is going to hit this nail on the head.
>> I'm asking you.
they were not adequately prepared to do the jobs they they were assigned to do.
>> Great.
Thank you very much.
All right.
Uh that concludes uh our first portion of uh presentations and uh board member questions.
Um and we are going to go to a 15 minute break.
Uh if we could come back at uh 12.
Uh well, we'll come back at 12.
We'll start again at 12.
Thank you.
Uh I will give a a couple of minutes because I want I certainly want the families to be able to get back uh and hear everything.
So, we'll give it a few more minutes for everybody to come in and take a seat.
I know sometimes there are bathroom lines, so that could be going on as well.
All right, welcome back.
We're ready for uh the next round of staff presentation.
So, I don't know if I defer to Mr.
Wagner or tell Bryce to take it away.
Mr.
Wagner.
The floor is yours.
>> Good afternoon, Chairwoman Hammedi, Member Graham, and member Imman.
My presentation will begin with background information on automatic dependent surveillance broadcast or ADSB and the airborne collision avoidance system or AAS.
Then I will discuss the limitations of the airplane's collision avoidance system.
Finally, I will present new collision avoidance technology for both commercial airplanes and rotocraft.
Automatic dependent surveillance broadcast or ADSB is a surveillance technology that uses GPS, aircraft avionics, and a network of ground stations to determine an aircraft's location with more precision than legacy radar technology.
ADSB includes two different services.
ADSB out and ADSB in.
ADSB out works by broadcasting information from an aircraft at least once per second regarding the aircraft's GPS location, velocity vector, altitude, ground speed, and other data to ground stations and other ADSBIN equipped aircraft.
Since January 1st, 2020, ADSB out has been required on all aircraft in most controlled airspace within the National Airspace System.
ADSBN is the capability to receive ADSB information.
This information can be received via air-to-air transmissions from nearby aircraft that are broadcasting ADSB out or via groundto-air transmissions from nearby ground stations.
Once an aircraft has ADSB in data, it can be used by various applications to provide the pilot with information such as displaying or alerting of nearby traffic.
The FAA does not mandate ADSB in.
The accident airplane was equipped with and transmitting ADSB out.
It was not equipped with ADSB in.
The airborne collision avoidance system or AAS is an onboard system that operates independently of ATC to reduce the risk of midair collision by displaying traffic information and alerting the pilots.
AAS works by monitoring the position and altitude of nearby aircraft that have an active mode C or S transponder.
Acast serves as a last resort safety net when other safety barriers have failed.
As shown in the figure on the right, AAS iterations include the traffic alert and collision avoidance system, TCAS 1, TCAST 2, and AAS X.
AAS products can issue two types of alerts to pilots.
Traffic advisories or TAS are aal and visual warnings used to alert the pilot of a potential collision threat.
Pilots can use a display to help them visually acquire the intruding traffic.
FAA and operator guidance instructs pilots not to maneuver based on TAS alone.
On the other hand, resolution advisories or RAS are aal and visual warnings with vertical maneuvers to avert a mid-air collision.
Only TCAST 2 and AAS X products issue RAS.
The vertical maneuver commands are issued to pilots to climb, descend, level off, or maintain their flight path to resolve traffic conflicts.
Pilots are trained to immediately obey RAS.
They serve as a backup to see and avoid the application of right-of-way rules and ATC separation services.
While TCAST 1 provides TAS, it does not provide RAS.
Since 1993, large transport category airplanes engaged in air carrier or commercial operations have been mandated to have TCAST 2 installed and active.
The airplane in this accident was equipped with TCAST 2.
As part of the investigation, a simplified TCAST display was simulated using the accident flight data.
Due to technological constraints, the simulation emits details that would have been available to the crew, such as the airport symbol, course overlay, extended runway center line, or terrain.
A symbol depicting the airplane is shown in the bottom center of the display.
This video shows the display set to a 5m range.
Five and 2.5 mile distance rings are visible.
An indication TA only located in the top right corner indicates that RAS are inhibited at this altitude.
The blue diamond depicts the relative position of the helicopter.
Below the blue diamond are numbers that indicate the helicopter's altitude below the airplane in hundreds of feet.
Occasionally, an up or down arrow will appear to the right of the blue diamond to indicate if the helicopter is climbing or descending.
Using accident flight data, a simulation of the accident airplane's TCAST display is shown on the right.
The blue diamond represents PAT 25 and the target symbology will change from a blue diamond to a yellow circle when a traffic advisory is issued.
The crew of flight 5342 received a TA for PAT 25 about 20 seconds before the collision.
>> Traffic traffic.
This is in accordance with the TCAST standards.
The airplane's TCAST did not issue an RA which would have commanded a maneuver to the airplane's crew to avoid the collision.
This functioned as designed per the TCAST standard.
TCAST 2 inhibits RAS below 1,100 ft above ground level when climbing and below 900 ft when descending.
Above these inhibit altitudes, a TA will always progress to an RA if the threat of collision increases.
During the development of the TCAST 2 minimum operational performance standards in the late 1980s, significant research and testing went into establishing the inhibit altitudes as a trade-off between maximizing the effective alerting envelope and minimizing nuisance alerts.
This trade-off was made based on the technological limitations available at the time of TCAST 2 development.
The accident airplane was below the RA inhibit altitude resulting in no RA being issued to avoid the collision.
A CAS X was developed as the next evolution of TCAST 2.
The goal was to create a system that improved existing collision avoidance alerting while reducing the number of nuisance alerts.
There are several variants of AAS X including AAS XA designed for commercial airplanes and AAS XR which is designed for rotocraft.
Advances since the development of TCAS 2 standards allow AAS XA to provide improved alerting among other enhancements.
AAS XA systems utilize ADSB in information in addition to transponder interrogations and replies.
AAS XA includes improved algorithms to more accurately reflect actual collision risk.
The AAS XA standards are approved by the FAA.
A series of simulations conducted by MIT Lincoln Laboratory using the circumstances of this accident showed that the crew of flight 5342 would have received a TA about 8 seconds earlier if the airplane had been equipped with AAS XA even though ADSB information directly from the helicopter was unavailable.
Staff have proposed two recommendations to the FAA in this area.
Although AAS XA can deliver earlier and more accurate alerts than TCAST 2, the current RA inhibit altitudes under AAS XA are the same as those of TCAST 2 and would have prevented AAS XA from issuing an RA under the accident circumstances.
Simulations of the accident conditions were performed wherein the airplane was equipped with AAS XA.
This time the RA inhibit altitudes were lowered to 300 ft above ground level.
The results of the simulations indicated that the risk of a near midair collision was with lowered RAS was reduced by more than 90%.
Staff have proposed two recommendations to the FAA in this area.
Other traffic display systems such as ADSB in displays can show directional traffic symbols that indicate a target's direction of flight, increasing pilots situation awareness and reducing the time pilots need to visually acquire traffic.
as depicted on the screen as a blue diamond.
Pilots must monitor the display over time to determine in which direction the target is moving.
Alternatively, as depicted on the screen as a blue arrow head, systems with directional traffic symbols rotate the symbol to indicate the target's direction of flight.
Most TCAST 1 and TCAST 2 systems display nondirectional traffic symbols.
Although AAS XA systems can display directional traffic symbols, they are not required for certification as the minimum operational performance standards are currently written.
staff have proposed one recommendation to the FAA in this area.
Another feature that could reduce the time to visually acquire target aircraft is the detailed traffic callouts in the ADSB traffic advisory system or ATAS.
A task generates verbal alerts indicating the clock position, relative altitude, range, and vertical tendency of approximate traffic.
For example, traffic 2:00 high 2 miles descending.
Acast does not include detailed traffic callouts that could reduce pilot's time to acquire a target.
Staff have proposed one recommendation to the FAA in this area.
The accident helicopter was not equipped with TCAS or an integrated ADSBIN traffic display nor was it required to be.
AAS XR the version of AAS X designed for rotoiccraft is still under development.
Therefore, no AAS XR systems are commercially available.
The simulations using the circumstances of this accident showed that had the helicopter been equipped with AAS XR, the risk of a near midair collision would have been reduced by more than 50%.
without changing the airplane's TCAST 2 inhibit altitudes.
Staff have proposed one recommendation to the FAA and one recommendation to RTCA in this area.
This concludes my presentation.
Now, Captain McKenna will present.
>> Good afternoon.
My presentation will discuss US Army helicopter operations and specifically ADSB out policies, transponder issues, UH60 Lima altimeter errors and night vision goggle characteristics.
FAA regulations that require ADSB out have exemptions for certain aircraft performing sensitive government missions where transmitting ADSB could compromise operational security.
The Army Aviation Brigade Standard Operating Procedures SOP specifically states that air crews conducting sensitive or classified operations shall emit transponder modes 3A and C in lie of ADSB out.
Trans transponder modes 3A and C reply to radar interrogations with an aircraft assigned squat code and altitude making it visible to FAA radars.
The SOP also prevented air crews from switching transponder modes in flight which prevented them from switching from modes 3A and C to ADSB out once sensitive portions of the flight were completed.
PAT 25 was not transmitting ADSB out during the accident flight, but had an active transponder.
The lack of ADSBout information was not a factor because the transponder replies provided the same altitude information to ATC.
A review of ADSB data from the UH60 Lima helicopters at the battalion showed that the accident helicopter as well as seven others did not have a recent history of transmitting ADSB out information since 2023.
We found that an incorrect transponder setting on these eight helicopters was the cause of the ADSB out issue.
The Army Aviation Brigade standard operating procedure restricted flight crews from changing transponder modes in flight.
This restriction prevented flight crews from transmitting ADSB out when flying nonsensitive portions of their mission.
Additionally, the US Army did not require a a recurrent check of the installed transponder to ensure it was transmitting properly, including the ADSB out function.
Because of this, the battalion was unaware of the ADSB out issue on aid of their helicopters.
Staff have proposed one recommendation to the US Army and two recommendations to the Department of War in these areas.
The Baltimore Washington helicopter route chart provided recommended paths without defined lateral boundaries.
As seen in the figure, the chart provided conflicting information about whether published altitudes were recommended or maximum.
This created the potential for pilot confusion about interpretation of the published altitudes.
Although some army pilots believed momentary exceedences above the published altitudes were acceptable, most regard them as maximum altitudes.
However, most Army pilots interviewed incorrectly presumed complying with the published altitudes would provide them separation from arriving and departing aircraft.
They rarely had the opportunity to test this presumption because they rarely encountered airplanes circling runway 33 and when they did, ATC typically instructed them to hold for arriving traffic.
Staff have proposed one recommendation to the US Army in this area.
Aircraft pedtostatic systems and barometric altimeters have defined performance specifications including tolerances for allowable errors after manufacturer.
These allowable errors can generally be split up into two categories.
Instrument error related to the performance of the barometric altimeter itself and position error due to the effects on the pedoatic system such as external aerodynamic effects.
Instrument and position errors can become additive but still must remain within their design criteria.
Analysis of flight data from this accident showed that PAT 25's pressure alim pressure altitude parameter was about 100 ft lower than other recorded altitudes.
Data from flights conducted on other UH60 Lima helicopters at the battalion found similar offsets between indicated altitude and true altitude.
The crew of PAT25 was trained to fly the helicopter routes using the barometric altimeter, including route 4, where the altitude ceiling went down to as low as 200 ft.
The crew of PAT 25 likely saw an indicated altitude that was 100 ft lower than the helicopter's true altitude and believed they were at or below the prescribed altitude ceiling for route 4 when they were actually about 100 ft above the altitude ceiling.
The US Army UH60 Lima operator's manual did not contain information on possible indicated altitude errors, such as how the installation of the external wing tanks can increase position error.
This information would be crucial for flight crews when flying on helicopter routes with lowaltitude ceilings such as the 200t altitude ceiling of Route 4.
Staff have proposed one recommendation to the US Army in this area.
Night vision goggles or NVGs amplify low ambient light levels to provide enhanced vision at night.
As with all technology, MVGs have advantages and disadvantages.
Helicopter operators generally consider the advantages to outweigh the disadvantages.
The advantages include improved nighttime visual acuity of low illuminated objects where normal night adjusted vision acuity is greatly reduced along with the ability to identify objects, obstacles, and other aircraft at a distance.
To illustrate the difference between aided and unaded night vision, we can take a look at these three images.
The images were taken about the same time from the same location looking south from Hannes Point.
The image on the left is normal unaded night vision.
Images in the center and on the right are looking in the same direction using green and white night vision goggles.
In the NVG8 images, stars can be seen along with aircraft approaching DCA as far as 14 miles, which are distinct points of light against the sky background.
The water features and lights along the distant shoreline are clearly discernible.
The disadvantages of NVGs include a field of view limited to 40°.
A monochromatic image, depth perception is degraded, both ground and airborne lights appear similar, light sources tend to have halos, and aircraft silhouettes are faint, making it difficult to determine type and size of aircraft.
The pilots are trained to compensate for the narrow 40deree field of view by using a focused visual search in sectors by turning their head.
Using NVGs along the low-level helicopter routes of route one and route 4 allows for better visual detection of obstacles such as cranes and towers which are located along these routes.
NVGs allow for the detection of aircraft above the horizon at a distance.
However, detection of other aircraft below the horizon surrounded by cultural lighting is difficult, especially if relative motion of that target is negligible.
Determining the other aircraft size, type, distance, and orientation is difficult because of the monochrome nature of the image, degraded depth perception, halos around point light sources, and the difficulty in perceiving the aircraft's silhouette.
All three crew members of PAT25 were using NVGs at the time of the accident and had four NG NVG sets on board, two green and two white.
However, the investigation could not determine which crew member was wearing or using which color of MVGs.
Thank you.
Dr.
Bramble will continue with his presentation.
>> Good afternoon.
In this presentation, I will define visual separation and see and avoid and I will discuss factors affecting see and avoid for each flight crew.
Visual separation is a method for separating aircraft in terminal areas and Nroot airspace.
Visual separation can be employed when a flight crew sees another aircraft and upon instructions from a controller provides their own separation by maneuvering as needed to avoid it.
The FAA does not define minimum separation standards when visual separation is being utilized.
See and avoid is the means by which visual separation is achieved.
Seeing and avoiding other aircraft relies on a flight crew's visual perception, attentional bandwidth, and use of collision avoidance technology.
In this case, both flight crews ability to see and avoid was hampered by poor radio communications, perceptual limitations, workload, and the shortcomings of collision avoidance technology available on their aircraft.
The helicopter flight crew was additionally influenced by expectation bias and assumptions about procedural separation between Route 4 and arriving airplanes.
As the helicopter was crossing the tidal basin, the controller advised, "Pat 25 traffic just south of Wilson Bridge is a CRJ at 1,200 ft for runway 33." Several other aircraft were visible in the same area at that time.
NTSB night vision goggle observations conducted later from the roof of a building near the helicopter's position, illustrated by the image on this slide, indicate the helicopter crew was likely unable to visually identify the CRJ in question because it was surrounded by similar looking targets.
This slide shows the helicopter's flight path in green.
The Mount Vernon visual approach to runway one in yellow and the airplane circling approach in blue.
Helicopter CVR recordings indicate that the controller's statement that the CRJ was circling was inaudible to the helicopter crew due to static interference.
The crew's workload was high at the time, and hearing the word circling might have helped them better anticipate the CRJ's subsequent flight path.
As the helicopter traveled down the Ptoac River, perceptual limitations subsequently reduced the ability of the crew to notice the airplane as a potential traffic conflict.
These included the narrow field of view provided by the crew's night vision goggles, a window post that intermittently obscured the crew's view of the airplane, a complex array of adjacent city lights, workload, divided attention, and lack of apparent motion as the airplane converged.
12th Aviation Battalion pilots carried electronic tablets that were wire wirelessly connected to a portable Stratus ADSB receiver that provided ADSBIN capability as shown in the yellow circle on this figure.
The tablets ran a mobile application called Forflight that was capable of displaying traffic symbols overlaid on a VFR sectional chart.
Pilots typically strapped these tablets to one thigh while flying the helicopter to assist with navigation and traffic awareness.
However, viewing the tablets required pilots to tilt their heads down away from the outside view and the flight instruments as shown in this figure.
Army pilots said they did not reference these tablets when flying low-level on the DC helicopter routes due to high workload and the need to frequently reference both the flight instruments and the outside visual scene.
Therefore, it is likely the pilots of PAT 25 did not reference the tablets when flying route 4.
The ForeFlight mobile application was capable of providing yellow traffic caution and red traffic warning visual alerts as shown on this slide with accompanying oral alerts.
However, visual alerts would have gone unseen without a visual inspection of the tablet and oral alerts likely would have been inaudible due to cockpit noise and their lack of integration with the helicopter's intercom system.
Further, it's unknown whether the tablets were turned on with the ForeFlight application open and configured for alerting at the time of the accident.
An integrated collision avoidance system with audible oral alerting and a traffic display within the pilot's instrument scan could have drawn the pilot's attention to the developing conflict and assisted them in locating the conflicting airplane.
staff has proposed one recommendation to the FAA and one recommendation to the Department of War in this area.
Radio communications came into play again when the controller asked if Pat25 had the CRJ in sight.
The controller did not provide a clock direction or distance.
As a result, the helicopter crew did not know where to look for the target.
As shown in this figure, there were multiple other airplanes in view that could have been mistaken for the CRJ.
The local controller directed the helicopter crew, Pat 25, pass behind the CRJ.
However, the crew was unable to hear the words 25 pass behindthe due to a partially blocked audio transmission.
As a result, the message the helicopter crew received was Pat CRJ.
The instructor pilot responded that the crew had an aircraft in sight and requested visual separation.
The controller, unaware that his transmission had been partially blocked, approved the request.
Hearing the instruction to pass behind the CRJ might have prompted the helicopter crew to ask a question of the controller or perform a broader visual search.
Staff is concerned that blocked transmissions can lead to miscommunications and reduce flight crew and controller awareness of potential hazards.
As a result, staff have proposed one recommendation to the FAA and two to the Department of War in this area.
The helicopter crews communications and actions leading up to the collision indicated that they lacked an accurate mental model of the CRJ's location and intended flight path.
As a result, baseline expectations likely drove the instructor pilot's visual search.
Only 5 to 7% of northbound arrivals typically landed on runway 33.
Therefore, the pilots likely assumed the CRJ referenced by the controller was one of several airplanes approaching runway one.
The instructor pilot's comment kind of come left for me reinforces the idea that he thought one of the airplanes approaching runway one was the airplane in question because although deviating to the left would have increased the distance from those airplanes, it would not have alleviated the conflict with flight 5342.
Moreover, neither the helicopter pilot nor the instructor executed an avoidance maneuver or vocalized any concern prior to impact, indicating they were unaware of flight 5342 approaching from the left.
I will now discuss factors affecting the airplane crews avail ability to see and avoid.
The local controller never provided a traffic advisory about PAT 25 to the crew of 5342.
The use of separate frequencies by airplanes and helicopters prevented the airplane flight crew from hearing the helicopter's radio transmissions.
This likely reduced the crew's awareness of the helicopter's location and intentions and their ability to anticipate a potential conflict.
FAA guidance on pilot response to TCAST traffic advisories, which had been adopted by PSA, stated that pilots should respond to a traffic advisory by attempting to establish visual contact with the intruder aircraft.
Pilots were explicitly prohibited from maneuvering based solely on a traffic advisory.
FAA guidance on response to TCAST resolution advisories, which had also been adopted by PSA, stated that the pilot flying should respond immediately to resolution advisory displays and maneuver as indicated, even if doing so conflicted with their air traffic control instructions, unless such action would compromise the safety of flight operations.
When an aircraft is below resolution advisory inhibit altitudes, TAS automatically switches to TA only mode.
PSA's flight operations manual listed TCAST inhibit altitudes.
However, when five PSA pilots were asked about the inhibit altitudes during investigative interviews, only one was able to correctly state the inhibit altitude that applied to the accident scenario.
Flight crew workload was high at the time the traffic advisory occurred, resulting in limited capacity to look for and acquire the conflicting traffic, and the pilots did not see the helicopter until it was too late to avoid.
The pilots likely would have performed an avoidance maneuver if a resolution advisory had occurred.
However, no resolution advisory occurred, and the crew was likely unaware of the severity of the developing conflict until just before impact.
The captain was manually flying the airplane and turning left onto final approach for runway 33 and decelerating when the traffic advisory occurred.
This required careful control of the airplane's lateral path, thrust, air speed, and vertical path.
Doing so required the captain to reference the flight instruments as well as the runway 33 edge lights and a precision approach path indicator or papy on the ground.
The captain likely did not have the spare capacity to perform an extensive visual search for conflicting traffic at that time.
As pilot monitoring, the first officer was also responsible for ensuring the approach remained stable.
Doing this required him to reference the flight instruments, the runway 33 edge lights and papy.
In addition, the first officer was likely monitoring the airplane departing runway 1 to ensure it would not conflict with their landing on the intersecting runway.
Thus, the first officer's capacity to search for potentially conflicting traffic was also constrained.
The ability of the airplane flight crew to notice the approaching helicopter in the final seconds while engaged in these primary tasks was further degraded by the helicopter's low conspicuity, complex background of cultural lights, and by the helicopter's lack of apparent motion in the visual scene.
This concludes my presentation.
staff is now ready to answer your questions.
>> Thank you very much.
We will now move to uh board member questions.
We'll start with member Graham.
>> Thank you, chair.
Uh I want to talk uh pick up with um visual separation as it applies to uh the helicopter operator.
It was clear from uh the interviews we did and the investigative hearing and other information uncovered during the course of this investigation that all helicopter operators in the DCA class B airspace had come to expect that they would request visual separation from DCA tower almost immediately.
Is that correct? >> Yes, it is.
>> That's correct.
Uh and that I want to iterate this point.
It it wasn't just army.
It was all the civilian operators also with according to Mr.
Dresser's uh uh uh witness statements in the hearing.
That is correct.
>> Okay.
Uh was it also true that in an interview with the Army standardization instructor pilot post accident, he stated that he and other pilots he knew would request visual separation from track traffic that they had not yet visually acquired.
Yes.
Are pilots allowed to request visual separation from traffic they have not positively identified and acquired according to FAA regulations? >> No.
>> No.
I'm kind of amazed by that comment in the interview.
Uh you being a longtime pilot as myself and I think Cap uh Captain Stone in the back longtime pilot.
um when you're given a traffic call, what what are you to do and what is your response to air traffic control? >> Yeah, once you receive a traffic call, then you search for the traffic and you identify it.
If you can't identify it, you tell uh traffic that you have negative contact or that you do have visual contact.
>> Okay.
Would you agree with that, Captain Stone? >> Absolutely.
>> Absolutely.
Yeah.
And if you're still looking for it, you maybe make some comment like that that I'm looking for the traffic.
But once you identify it and then you ask for visual separation, are you required at that point to keep your eye on that traffic? >> Absolutely.
>> And if you lose the traffic at any point, what is your what should you do? >> You'll have to know notify ATC that you have uh lost the traffic.
Yeah.
Traffic is no longer in sight.
>> Would you agree with that, Captain Stone? >> Yes, that is correct.
>> Did that happen in this case? No, it did not.
>> Did not.
No.
>> I like that you pointed out the fact that it was be pretty pretty much impossible to pick out which one the traffic was uh in your uh visual study out there.
Um now uh 15 seconds prior to the collision the uh helicopter IP had uh they were about a mile away from 5342 a requested visual separation.
Uh is that correct? >> I believe so.
Yes.
Do you believe or the rest of the team believe that they had sight of the correct aircraft at that point? >> We don't believe he had sight of the correct aircraft.
>> Yeah, I would agree.
and unfortunate for them.
Um, do you think if they, and this is an opinion question, if they had a side of that aircraft where it was, they would have made some kind of positive control movement to um, deconlict with the aircraft.
>> Absolutely.
>> Had an appropriate safety alert from the air traffic controller, would would that have helped in this situation? I believe it would have because it would have directed their um direction.
They needed to look for the traffic.
So, they would have looked in the one 1 to 2:00 position for the traffic and they would have possibly identified it.
>> I agree.
Do you feel I'm going to ask this of Dr.
Bramble.
Um, do you feel the lack of the circling uh call that they did not hear the helicopter crew? Do you think that played into them not being able to visually acquire the aircraft? >> I do think it had an impact.
Yes.
>> Now, they they did, as far as we can tell, did hopefully hear 33, but do you think how do you think that played into this at all? So they were high workload at the time calling out cranes in the middle of a turn.
The instructor was giving the pilot feedback on use of pedal and um uh you know they heard this partial transmission that said the CRJ was going to 33.
Realizing that was going to conflict with their flight path further on on runway on route 4 required them to do a fair bit of thinking about the transmission.
And um it appears that they didn't process that information very deeply.
Just thought that target is a long ways away.
We'll worry about it later.
Um if they had heard the word circling, that would have been a salient cue that something unusual was going on here that they might need to keep an eye out for.
>> We'll talk more about this when I come back.
I'm done.
Thanks.
>> Thank you.
Member Amen.
>> I just want to establish something I guess is a little bit of a baseline.
Uh Bryce, uh we have what have we established the what we believe to be the exact altitude at which this crash occurred? >> Yes, I think we have.
I think we believe it came together at 278 ft.
Is that correct, Miss Dia? >> Yes.
>> Okay.
I'm just wanted to try to get that Captain McKenna, I want to go back to your actually your presentation, your deck, uh and it was on page I just want to clarify something.
Uh it was on page 25 that we stated PAT's 25 PAT 25's pressure altitude was 100 feet lower than other recorded altitude data.
But then on the next slide we say on page slide 26 likely believe they were likely indicated altitude 100 ft lower than true.
Do we have a difference between what we verified and what we know to be absolute >> between the absolute altitude of what we what the aircraft was at versus what we believe they thought they were at? >> I'm sorry.
One slide says that it was 100 ft lower.
The next slide says likely was 100 ft lower.
It might just be a statement, but the FDR recorded pressure altitude, if you made the correction for the local barometric conditions, was 100 ft lower than the other altitudes.
And we believe that that is likely what they saw on their altar.
>> Okay.
I'm just trying to get to that to that area because it's whenever we say likely and was sometimes that can be a little bit different.
Um so we indicated at least in the airworthiness report that flight testing by Sakorski found that the installation of the external store systems uh would increase this position error meaning that would that would change possibly what the altitude would be.
Is that correct? >> Yes.
uh when you install the uh the wing tanks, it changes the pressure dynamics around the outside of the aircraft and can lower the uh recorded pressure altitude.
>> So, is there any way for a crew to mitigate that or do they have to just take it into account and know that it's there? >> Um I would pass this to Mr.
Shin >> Shun for the uh UH60L.
Uh the crew would have to take that into account um mentally in their mind.
So, is there anything in there at least pre-flight or training or manuals whenever they know they're going to have stores that they would pre-brief or at least indicate that or make it aware between the crew? >> That that information was not in the UH60L operator's manual.
>> So, they just kind of have to know.
>> Uh, we don't believe they knew since the army did not have that published in the manual.
>> Okay.
and we have a recommendation in that area.
>> I understand.
I'm just trying to build into some of the as we said before there's not individuals.
This is a systemic issue with a lot of manuals and a lot of policies.
Uh how do they educate Army air crews on the position error when they will encounter with the CSS? Is that in their flight training? Is it passed down? Do we have do we get any information of why how flight crews would know this? >> They wouldn't know it.
It's not in their manual.
It's not in the training.
>> Okay.
So, in the helic uh I'll go on to a different area real quick.
So we uh we did some discussions in regard to uh ADSB and the alerts and I guess if the helicopter So let me go back.
If a an ADSB display indicates a traffic advisory or an RA, does that override a controller's instruction as a pilot? >> Yes.
But earlier in the last panel, we we showed where it needed to be an emergency or it had to be something in regard to the FAA circular specifically what the controllers are told.
And no pilot in command may deviate from that unless an amended clearance is obtained or emergency exist or the deviation is a response to a traffic alert and collision avoidance system resolution advisory.
Yes, if it isn't if it's coming from an Acast system and it's a resolution advisory, then the pilot is authorized to deviate from the ATC clearance by the F that you had up on the screen earlier.
>> So, I guess my question is if we ask them to lower the RA threshold from the 900 ft now, does that mean that all air traffic controllers will have to be retrained as as will pilots? >> There will be training.
Yes.
>> I'm just curious if that injects additional risk.
>> There there's always some risk associated with that, but I think the staff has sought this through.
I think there's much more benefit to it than risk.
>> I'm not disagreeing with your recommendation.
So, only think that I'm just trying to make sure we get all the facts out there.
>> Thank you.
Good thing we have an FAA that will when doing rule making determine any risk and address that risk including training needs.
Um when you are uh uh flying an MSL route uh Mr.
McKenna, would you be using your barometric altimeter? >> Yes, >> that's what you'd be referring to.
>> That's correct.
>> And uh could the left seat pilot have easily observed altitude indications on the opposite side of the cockpit? But It'll be a little difficult but they can they can see it.
>> Looks a little difficult from >> across cockpit.
>> Yes.
Yes.
We actually did an observation in the simulator.
It was very difficult to see the instruments on the opposite side because they're inset and you'd be looking from the side.
>> Correct.
So 17 seconds before collision, the aircraft were about 84 nautical miles apart.
The hel helicopter's recorded radio altitude was 281 ft.
They would have thought given our uh assessment that the barometric altimeters they thought they were 100 ft lower.
They would have thought they were below 200 feet.
Correct.
>> Correct.
>> 6 seconds before collision it was 266 feet.
Again they would have thought they were under 200 ft.
>> That is correct.
4 seconds before collision 278 ft.
Again, they would have thought they were under 200 feet.
They would have not known.
>> That's correct.
>> Okay.
So, um we have established that they thought that they were below 200 ft.
We conducted three flight uh demonstration flights.
Uh all three showed barometric altimeter readings that were 80 to 130 ft lower than true altitude.
Uh we also looked at uh the ESS which would add an additional 30 ft.
Um were the instrument and so we've got instrument errors, we've got position errors and then you from from roer downwash, right? And then we've got uh some uh adjustments for temperature on the day.
Uh did were the instrument and position errors known by the army? >> Army, not the pilots? >> Yes.
>> How about the pilots? >> No.
>> Should they have been? >> Yes.
>> Would it have made a difference if they actually knew they were below 200 ft? I believe it would they would incorporate that into their training and then they would fly a more appropriate MSL altitude on those routes.
>> Yeah.
So you have this tolerance stack of uh a bunch of instrument position errors.
Uh you've got temperature correction which all add up to this 100 ft.
Then you've got the addition of 30 for the ESS.
Um what is the Army doing to address this? They're adding that information into their flight manual for the pilots.
>> Are they going to not just have it in writing on a manual but actually train their pilots accordingly on what to do there? >> I believe so.
Yes.
>> Yeah.
Okay.
So 100 ft difference may not make a difference in Afghanistan, but it certainly makes a difference when you're on helicopter route 4 crossing uh um uh civilian aircraft on approach to landing to runway 33 where there's only 75 ft of vertical separation.
There's zero buffer, nothing.
Yeah.
Um Dr.
Bramble, can you talk a little bit about your expertise? You're a national resource specialist for the NTSB.
Talk about your expertise in human performance and some of your background.
>> Um, sure.
I have a PhD in human factors from an interdisciplinary uh program in psychology and industrial engineering.
And I worked for a flight safety um a flight training and simulation company for a few years.
I've been at the NTSB for 27 years investigating major accidents for the last 24 years and I chair I chaired the um human factors working group for the international society of air safety investigators and I currently chair the government specialist subcommittee uh with other human factors investigators from 10 other countries.
Do you believe the limitations of see and avoid which we've referenced back to 1968 in numerous investigations 96 alone within the past two decades uh could be overcome simply by uh telling a pilot uh to do better pilot diligence in scanning for traffic? >> No, we need to do a better job supporting pilots in that task.
>> And I'll come back to that.
Member Graham.
Thank you, chair.
I want to go back to uh trying to figure out why Pat 25 crew didn't positively identify uh flight 5342.
Dr.
Bramble, can you um explain what expectation bias is? >> Sure.
Um, it's sort of related to confirmation bias in that when you have a strong expectation about things going a certain way, you tend to notice confirming information and overlook disisconfirming information.
And is expectation bias a well-known vulnerability in human performance? >> Yes, it is.
>> Yes, we see it a lot, don't we? >> Yes.
Captain McKenna, uh, did Army 12th Aviation Battalion flight crews receive any specialized training on the route structure, including approach and departure paths at DCA? They did not receive specialized training.
The training they received was uh general aerial familiarization and of course their basic uh airmanship.
They'd recognize, you know, runway alignments and and traffic in that way, but there is nothing specific about DCA.
>> Okay.
So would it be fair to say that their expectations for where traffic would be arriving and departing from at DCA would be driven by their experience? >> Yes.
>> Okay.
Rather than formalized training.
Correct.
>> Correct.
>> Okay.
As we uh pointed out in that table I had up earlier, table 18, we see that about 90% of that traffic, even at night going on a north flow, landing runway one, majority of that traffic was landing on runway one.
So with their prior experience, I would say the majority of the time, if not almost all of it, they would have been seeing traffic landing on runway one and not on 33.
Would that be correct? >> Yes, that's correct.
And that's what we heard in their statements also when we interviewed them.
>> So Dr.
Bramble, would that lead to some expectation bias? Maybe.
>> Certainly.
That's perfect setup for it.
>> Yeah.
I mean, I I might have a little theory here and be speculation, but I don't think that hearing anything on 33 or the circling before 33 would have made a difference in this case, unfortunately, knowing what expectation bias is.
And if you don't have that traffic initially, but you're looking at a a line of lights that may led them to continue to look towards runway one immediately.
just it's a theory, but we don't know and we'll never know.
Let me move on to uh Mr.
Wagner.
Um I want to talk a little bit about AAS XR.
Um what how how has AAS been adapted for the rotor wing uh aircraft? Acast XR is different from previous iterations of AASS in that um instead of just offering vertical maneuvers to air crew, it offers also horizontal maneuvers.
And additionally um the committee that's writing the minimum operational performance standards is looking at um having significantly lower inhibit altitudes for XR when compared to XA or XO for instance.
>> Okay.
And you said it's commercially it's not available at this point.
Correct.
>> It's still in development.
That's correct.
>> What are minimum operating performance st standards or MOPS? So minimum operational performance standards, MOPS, those are um basically a set of criteria that an organization like RTCA develops to say um what a company would need to verify that they that their product meets those those standards in order to get a certified product.
Have those standards been developed yet for AAS XR? >> They are in development currently.
>> Okay.
So my concern is the simulation that we run what what performance standards are used.
Is it these ones that they're they're uh maybe they're thinking of having or what what do we know? What are what are the performance standards for the simulation? So the simulation used performance standards that they are still in draft.
There's um some ability to modify those.
Um however, in speaking with the committee chair.
Um just a week ago, he stated that they intend to have the standards finalized by December 2026.
So within less than a year from now.
Okay.
>> So we believe that it's mature enough technology that we can make recommendations in those areas.
>> So the simulation software is a best guess at this point of how it's going to be the standards are going to be.
>> Um I wouldn't characterize it as a best guess personally.
Um, I think that there's still a couple small things that they need to validate before they finalize the standard.
>> Okay.
>> Um, I think Captain Stone may be able to offer a little bit more information.
>> Be quick.
I'm past my time.
>> No, I I would say that the standards are 95 to 98% complete.
So, they're much more than just in development.
>> Okay.
And how long have they been developing this for? >> A long time.
>> Long time.
Okay.
Thank you.
Member Inman, >> I want to go back to this.
This was actually a um an evaluation flight.
So, we know that that was there's evaluation going on.
And I'm I'm kind of concerned or wanted to figure out maybe this is human factors or actually an expectation bias.
But we noted 27 mentions of wind and turbulence, a brief loss of the crew coordination on who was flying, a failed landing, and a pilot initiated wrong turn.
All during this evaluation flight, at what point does an instructor pilot say, "We need to knock it off.
It's time to end." >> Well, this was a a proficiency evaluation.
You're correct.
And so, uh, the pilots being evaluated on their ability to, um, fly the aircraft, manage cockpit resources, navigate, whatnot.
Um, I think as an instructor, pilot doing an evaluation, don't really uh, you're looking for an unsafe action or situation where you would call it, knock it off.
Otherwise, I think most instructors just continue with the flight.
is do you think instructor pilots are given enough latitude or at least direction on when they should knock it off? >> Yes, that's totally their decision.
>> I know.
But are they given enough training in that regard? >> I believe so.
They're especially u selected because they have good judgment abilities.
>> Do you think that this IP was empowered to uh to end it despite uh evaluating a senior ranking official? >> Absolutely.
He was uh the um instructor pilot in command.
>> Okay.
Um in their pre-flight risk assessment, do they take into account how many hours a pilot has recently flown? >> Yes, they do.
>> And how is that weighted? >> They use a u um risk assessment tool called an ARCOP.
And then it's automat that information is automatically integrated into the risk assessment tool and a uh risk rating is generated.
It's based on their how many flight hours they have, the recency and what modes of flight they've flown and whatnot.
So it's uh automatically generated >> and I think we we've heard that all the crew members aboard both aircraft were qualified to perform their duties.
Um, but is is there an attempt or a bias in order to try to know that we've got a deadline to get this done? We need to get it done regardless of what happened.
>> Uh, we do not, the investigation did not reveal anything like that.
>> Okay.
Okay.
Um this is just for my awareness too and it was interesting whenever we unfortunately we we do listen to the CVRs uh to make sure that we're getting the tone and tenure and I've always heard of the sterile cockpit rule that applies under 10,000 ft for commercial aviation.
Can you describe that? >> Maybe Cap Stone can.
He's a airline pilot.
>> Yeah.
So it's below 10,000 usually it's 10,000 ft.
Some airlines have it up to 18,000 ft and it's basically only operationally significant information can be discussed.
So you can't have a conversation about where we're going to go eat tonight or anything like that.
So it's got to be directly related to the safe operation of that flight.
>> Does the same thing apply to military or to helicopters? >> Uh not with helicopters.
No.
>> You're operating below 10,000 feet almost all the time.
So, it's a different risk for a helicopter than it is for a fixed wing airplane.
>> Different risk.
>> They're treated differently, I should say.
>> That's correct.
You know, uh, helicopter operations are typically low altitude.
You're always going to be communicating with your crew on the, you know, situation of the aircraft, looking for traffic, navigating, and, uh, you know, communicating with each other about what's going on.
So yeah, that's always going on in helicopter.
>> Okay.
Um, and I know we talked a little bit earlier about the uh the bias that these pilots had not been seeing uh aircraft arriving on 33 very often.
Um, and did we have a number of ATSAP reports on that? Did we find where this was specifically more defined at night? Do we have any data on that? >> I think that's something that we'll be covering in the next panel.
>> Okay.
I'll wait for the next round.
>> Thank you.
Um, this may be for Mr.
Ho.
>> Hi, Nathan.
Welcome to the discussion.
All right.
Uh, so I referenced in my opening statement a 1968 midair collision and again in ' 69, which was the worst air disaster in Indiana and remains.
So, we just passed uh not too many not a few years ago the 50th um uh the mark of the 50 years since that occurred.
And I mentioned the board's significant work in 1968 and 69 to hold midair collision public hearings in general and issue a number of recommendations including again after 1968 recommending the expeditious development of a collision avoidance system and implementation on all civil aircraft.
Um, n also of note in July of 1969 was the NTSB's first ever recommendation on another collision avoidance technology.
positive train control to prevent trainto train collisions and to prevent incursions is basically a train slamming right into a work zone and hitting people.
We issued a recommendation on positive train control so numerous over the decades over 50 years that I could hardly count.
It was many, many times, sometimes six times a year.
For 50 years, we issued that recommendation.
In 1969, there was no standard for positive train control.
There was no group that got together and agreed on a written document of a standard for positive train control.
It didn't even exist.
Did not exist.
But we recommended it.
And guess what? Yes, it took us 50 years to get there.
But Congress mandated it.
They mandated it because of our recommendation.
But in between 1969 and when it was mandated and fully implemented in 2020, over 300 people died and 6,200 people injured, seriously injured, death after death after death.
It wasn't the first time we issued a recommendation where technology did not exist or there was no standard in writing.
Let's just review.
1968 when seat belts came into play.
We it issued our first seat belt recommendation.
Now knowing 450,000 lives were saved.
Think about the tremendous push back on seat belts at that time.
Nobody wanted to wear a seat belt or install them in a car, but we were right out there championing championing it.
1970 was issued a recommendation to rapidly shut down failed pipelines, now known as automatic and remote control shut off valves.
Didn't exist in 1970, but we investigated 15 something explosions, including one at a daycare that killed so many people.
So many people over the years.
Excess flow valves.
is another one to shut off distribution pipelines before a house exploded we kept after that one time and again an excess flow valve didn't even exist at the time ground proximity warning systems 1971 windshar detection technology 1973 car seats and passenger vehicles 1983 runway incursion technology or asdx 1990 direct pilot alerting We're still after it.
1990 automatic emergency braking, forward collision warning, 95 fuel inerting systems, 1996, electronic stability control, 2002, advanced speed limiting technology, 2011, connected vehicle technology, and thank you for member Graham for championing connected vehicle technology.
First recommendation 2012 didn't exist.
drug and alcohol detection technology, whether it's dads or driver monitoring, it didn't exist 2012.
If not us, then who? Every single time, we have championed technology to push for lives saved.
And when we issue a recommendation and FAA decides we can't do it this way, we need to find another one.
Mr.
Hoy, is there an option for FAA to say, well, we can't do it this way, but we are going to do it this way.
>> Yes.
If a recipient has another um alternative plan or solution for recommendation, they can uh propose it to us and we'll consider it and if we think it's appropriate, we will close it.
Acceptable alternate action.
>> Thank you very much.
Member Graham.
>> Thank you, Chair.
Um, I want to talk about technology now, too.
Um, I've been a big proponent of technology to help as a backup system out there.
Um, TAS was a big deal when it came out.
It was, it has matured over time to be a really good thing and it has saved my rear end a few times over the years with a resolution advisory.
Uh unfortunately in this situ situation, this accident it uh it had limitations where it wouldn't work or wouldn't be effective.
I think we we've gotten to the bottom of that.
So um I want to talk a little bit about AAS and ADSBN um going forward here.
I think it's important to point out um I'm a believer in ADSBN, but I'm only a believer if we do it right and how it's integrated into the into the cockpit.
Just to have it in is worthless, but if we do it right, what would that look like? cabin stone.
>> It would require that the there be directional targets and that it's in a in a location on the cockpit the in the primary field of view of the pilot so that they could see it and also the audio for the alert integrated into the uh their audio system so they could hear it.
>> Yeah.
So, I would agree with that.
You know, would what else would be good um tied in with that? Would it be directional, alert, range, altitude? >> Yes.
And one of our recommendations we actually were asking that uh you know more information be given to traffic alerts so that and that's the first step in the in this you know in developing situational awareness as to where where the traffic is where the conflict is and we're recommending that that that be enhanced beyond what it is today.
>> I would agree.
We've seen in the past just to have ads be in and catch a can for our IIC.
Mr.
for banning that just having it without an alert, a oral alert or a visual alert was useless, was it not? And unfortunate.
So, um let's talk a little bit about AAS.
Um is AAS does it incorporate ADSB in? >> Yes, it does.
>> Okay.
So is that just one of the uh inputs to Acast? Would that be correct? >> Yes, it Acast can function off of inter and reply just like TCAST does, but it has the ability to actually look at the geometry which TCAST really doesn't look at geometry.
So >> correct.
Correct.
Is there any iteration of AAS out there that would allow for the ADSBN to make the uh clock code range and altitude callouts? There is no Acast that's specified to do that today, but it could be >> could be might be neat if we're going to use it ADSB and with it.
It would be nice for the crew.
>> Um, so these systems are great, but they're not the primary means of traffic deconliction in most cases.
Is that correct? >> That is correct.
is primarily uh air traffic control separation and could be visual which could include visual separation.
>> Correct.
Yeah.
And in this case this unfortunately was uh an incorrect uh visual separation I would say on behalf of the tower and the air crew in this case.
And yet if we had some of this technology in the future maybe this accident doesn't happen.
But it was a breakdown initially in the visual separation.
>> I see my time's up.
Thank you, >> Member Inman.
>> I appreciate it, Chair, you bringing up the PTC and it actually resonates with me a lot.
Um, and I would go back to whenever I was the chief of staff the US Department of Transportation, um, 2020, uh, we actually got that implemented.
But it also reminds me that it was another recommendation and unfortunately from the Kogan air crash in 2009 that had been outstanding, the pilot record database that the NTSB had recommended and it took from 2009 until 2019 to get the FAA to adopt that.
And I I will say this, I'm being on the receiving end of that.
I'm going to speechify a little bit, but it's going to take everyone in this room and everyone that's watching.
I think we have a receptive administration, receptive secretary, a receptive FAA administrator, but you have to continually put that in front of them.
And that's how PTC, that's how PRD, the Kogan families became huge advocates.
And that's how they were successful in getting that done.
So, it's a good message to know that we don't give up, but no one can no one else can give up either.
So, I just wanted to mention that.
Um, let me go back to to see and avoid.
How are pilots trained differently to see and avoid versus a helicopter pilot versus a fixed wing pilot? >> Not particularly.
>> Is it not just basic knowledge that you don't hit something? Sorry.
>> Yes.
I mean, isn't that the core fundamental is don't hit something else? And yet we're using this terminology to see Rocky.
I see you back there laughing a little bit, but I know it's not funny, but it is it's it's so ludicrous.
It's kind of ironic, I think.
Am I wrong in that? >> I agree.
If >> I could add I mean, in order to avoid something, you have to first see it and see and avoid.
And I think we identified some challenges in this case for that.
I just wonder why we're continuing to let and I I don't disagree that that the technology can help avoid part of that.
Uh but I go back to we train people to see and avoid or request visual separation.
It's it seems like it's just meant for the sake of expediency rather than it is actual safety.
Is that a good assumption? I think staff determined in this case that uh there was an over reliance on it in order to ensure the ex the efficient flow of traffic in the terminal area.
>> So it's more about getting the planes efficiently where they need to be.
>> We believe there was a drift in operating practices uh due to that pressure.
Yeah, >> I think shouldn't it be getting them there alive versus efficiently? Yes.
Sounds a little rhetorical, I know.
Sorry.
Um, in regard to night vision goggles, is there ever a time whenever a helicopter crew would advise their air traffic control or they would let them know that we're using we're in a training either in a training flight, a proficiency flight or using night vision goggles.
Right now there's no requirement to do that, >> but we've identified increased risk.
Obviously, when helicopters are using night vision goggles and operating in a complex airspace, >> there's dis advantages and disadvantages in using night vision goggles.
and all the operators, all the helicopter operators in the National Capital Region use them um because they think it's advantageous, but it doesn't they advantageous because they're able to better discern oncoming traffic or because we've seen from our visualization, it significantly reduces your field of view.
>> Right.
for helicopter operations.
Um, you're flying at low altitude.
So, you're not just looking for other traffic, you're looking for ground obstacles and and other aircraft that are lower to the ground, other helicopters.
So, the night vision goggles um are a big advantage in that environment.
>> Dr.
Brandle, did you have something you want to add? I was just going to say, yeah, it operating without them would likely increase the risk for helicopter crews, but we have um made other recommendations that would reduce the risk of collision in the area, such as relocating the route so they don't have to spot oncoming airliners and increasing the use of collision avoidance technology.
>> Okay, >> I see I'm out of town.
>> Thank you.
Uh, Rocky, uh, does American Airlines have ADSB on 302 of their A321 aircraft? >> Yes, they do.
>> That was, uh, cockpit display.
>> And it has an cockpit display in the pilot's primary field of view.
Yes.
>> Yes.
I flew jump seat and was able to watch it in operation.
Tim, you're a pilot.
Do you have ADSBN? Yes, ma'am.
I do.
>> And you have a receiver that you bought.
It was a Garmin.
Probably a couple hundred dollars, right? >> I think about 400.
But yes.
>> And what does it provide you? >> So, uh, the interesting thing, I'm not flying a 121 airline.
>> That's right.
>> But, uh, for me, I do fly a 1931 biplane.
Okay.
So, this airplane has no electrical system whatsoever.
I have to turn the propeller by hand to start it.
But for $400, I bought an 80SBN unit that velcros to my panel.
And using my phone in my ear, if I get traffic, I can hear the direction of that traffic, how far it is away from me, a clock position, and its altitude.
>> Right.
Uh Olivia, can you pull up the um uh PowerPoint slide that I had said sent to you all on the 48 seconds? All right.
So, what we're looking at here on your right is had uh PAT 25 uh they had a four-flight uh do you have forflight, Mr.
Loaron? >> Yes, I do.
>> So, you have fourflight with ADSB in? Yes.
>> And so this had they gotten that alert and had it been integrated in the helicopter would have had 48 seconds before a collision to do something.
We know for a fact uh or we know from our analysis that they didn't realize that plane was there on the left.
There was nothing.
Um on the on the left here we have uh flight 5342 and we did a simulation.
And they're not even they're barely even I mean they're just circling now from the Wilson Bridge 59 seconds before collision with ADSB and they would have gotten an alert instead of 19 seconds.
Correct? >> Anyone? >> Yes, >> that is correct.
>> That is correct.
Uh so would you rather have 19 seconds or 59 seconds? Well, I'm not sure who the questions to, but I'd rather have the 59 seconds.
>> That's right.
That's right.
I mean, look how far out they are.
That is our simulation.
So let me tell you with respect to ADSB, the NTSB has recommended ADSB out and in uh supported airborne traffic advisory systems that include oral and visual alerting functions 17 times since 2006.
on part 121, part 135, and general aviation aircraft.
In 2008, the board took a position in response to an FAA notice of proposed rulemaking which required ADSB out.
Uh that the NTSB expressed concern that the FAA did not plan to require ADSB incapability.
We stated the safety board believes that the benefits of ADSB technology warrant rapid adoption.
We stated our concerns in that letter.
We said uh uh to FAA at the time said they had not identified as a requirement for maintaining safety and efficiency of NAS operations.
We told them we disagreed said their assessment was incorrect and that equippage of aircraft with ADSB incapability will provide an immediate and substantial contribution to safety especially during operations in and around airports.
Uh we went on to say ADSBN provides a ready method of sending warning information to pilots, providing confirmation of clearances on and on.
It is a backup because why humans make errors? We know this.
Why do you have technology? I mean, uh member Inman talked about, you know, they implemented positive train control.
Guess who wrote it in legislation? that'd be me.
Why? Because of your work over and over and over again.
And so, uh, we went on, uh, to continue, uh, recommending ADSB in.
I wonder where we would be today if they actually did it when ADSB out was required.
Where would we be today? Perhaps we wouldn't even be here.
59 seconds.
48 seconds.
We might not have to be here.
So, uh, my time is up, but I'm going to come back to this.
Member, uh, Graham, member.
Okay, just along those lines, I do want to stress when we're thinking about where we are today, I want to read the last two responses from FAA and it predates this administration.
Honestly, thank you to Secretary Duffy who has been a partner from day one, day one on this.
You know when the your first day on the job is a devastating tragedy that is stays with youth for the rest of your tenure if not your life.
And what did he do? He committed to ADSBN.
And the last administration June 26, 2023, a response from then acting administrator of the FAA Paulie Trottenberg was the FAA's air.
This is to us.
The FAA's air traffic organization and aviation safety organization met to discuss this recommendation and review all guidance and requirements associated at the time in response to Ketchacan parts 91 and part 135 and ADSB out and in.
We determined that our current ADSB requirements adequate adequately address the needs of aviation safety and will not pursue any additional ADSB requirements at this time.
And then uh we had a another one no the last uh from them was November 6, 2024 Mike Whitaker response to to us saying based on our discussions it has been determined that our current ADSB requirements continue to adequately address the needs of aviation safety.
The FAA will not pursue additional ADSB operator requirements at this time.
In other words, they were okay without and not in.
For those that are watching, that's like having a conversation with a wall in.
In order to communicate two entities out, you got to receive in.
You got to take that information in.
Now, in this case, uh uh you know, and I might be at five minutes.
Do you want to in this case I I I want to ask one thing uh about Tisby.
If this aircraft had ADSB in and we know that the army's helicopter had ADSB out would have been on but it was programmed incorrectly on installation so it was basically not transmitting out.
But even if I and I've seen comments, it it wouldn't have mattered.
I've seen this in the press.
It wouldn't have mattered if it had ADSB in because that aircraft, that helicopter, ADSB out wasn't working.
Do you agree with that? Am I going to ask Mr.
McKenna or am I going to ask Mr.
Chun? So the helicopter was not transmitting ADSB out but with the TISBY service that would provide um the helicopter's position to ADSB and displays.
>> So in other words even though the helicopter wasn't transmitting out if that aircraft had ADSBN it would have gotten information.
Is Tisby everywhere in the airspace? >> No.
>> No, it's not.
Which is why we still need I see the army in the audience.
This does not mean you don't get a pass on ADSB out.
The TSB isn't everywhere.
You have to have ADSB out and in.
Okay.
Um, anything further? Okay, we're going to take a 15minute break, lunch break, apparently.
I don't know what time it is.
And, uh, 1 hour.
We'll be back at 2:30.
All right.
Uh, welcome back.
Uh, we're now going to hear from Dr.
Jana Price.
on safety management systems and safety data.
Dr.
Price, >> thank you.
And good afternoon, Chairwoman Hamdy, member Graham, and member Enman.
My presentation will cover issues relating to safety data and safety management systems or SMS.
Specifically, I will describe data sources that could be used to measure midair collision risk.
I will also discuss who had access to those data and how they were used before the accident.
Finally, I will discuss SMS programs at PSA, the FAA air traffic organization or ATO, and the Army.
Within days of the midair collision, the Aviation Safety Information Analysis and Sharing System or ASA provided the NTSB with data concerning encounters between helicopters and commercial aircraft near DCA.
The initial reports indicated that there had been thousands of so-called close proximity events near DCA in the years before the accident.
In the weeks and months that followed, seemingly conflicting estimates of proximity events and near misses were shared by FAA.
As a result, the NTSB formed a safety data group with the goals of documenting safety data sources that could have provided indicators of midair collision risk before the accident, whether those data were available to stakeholders, and how they were used in the context of SMS safety assurance processes.
The data sources were grouped by whether they were comprised of safety occurrence reports or whether they represented objective position data.
This slide lists the safety occurrence reporting systems we reviewed.
The blocks outlined in yellow were managed by the FAA, including the Aviation Safety Action Program or ASAP, the air traffic safety action program or ATSAP, mandatory occurrence reports or MOS, and the near midair collision system or NEMAX.
The aviation safety reporting system or ASRS was managed by NASA and the army safety management information system or asmus was managed by the army.
The primary sources of the safety occurrence systems were pilots or air traffic controllers.
The systems varied in terms of whether reports were required or voluntary and whether those who reported voluntarily were deidentified and protected from punishment.
There was also variability in which stakeholders had access to the data.
Our review of safe of subjective safety occurrence reporting systems found that that more than 18 reports per year on average of close calls between airplanes and helicopters near DCA in the four years before the accident.
When available, we reviewed event descriptions and found reports describing issues like those found in this investigation, including airspace complexity, problems with ATC communications, challenges with combining controller positions, and helicopters flying above recommended altitudes.
This slide provides an overview of the position data systems we reviewed.
Again, the blocks outlined in yellow were managed by the FAA.
The primary sources of data were from aircraft and from surveillance radar.
The flight operations and quality assurance or FOQA and the ADSB systems provided information about TCAS resolution advisories and traffic advisories.
systems known as aviation risk identification and assessment or ARA and performance data and analysis report system or PERS provided objective information about aircraft proximities using radar and other surveillance data.
Data accessibility varied by source and although the FAA generally had access although the FAA generally had access to all data sources.
Our review showed evidence of numerous close encounters between airplanes and helicopters near DCA.
For example, TCAST resolution advisory data captured by groundbased receivers identified about 15 events per month on average within 10 nautical miles of DCA.
The Arya system found that airplanes and helicopters came within one nautical mile laterally and 400 ft vertically an average of 390 times per month.
Peters found an average of 65.6 six instances per month in which airplanes and helicopters came within 1,000 ft separation and 5.6 instances per month in which they came within 500 ft separation.
This heat map shown on this slide comes from Petars and shows encounters between arriving and departing airplanes and helicopters flying on routes 1 or four for the period January 2018 to February 2025.
Additional analyses of radar data found that nearly half of all helicopters that transited Route 4 went above altitude limitations at least once along the route.
However, we found that neither the FAA nor the Army identified these exceedences before the accident.
In all, we found that multiple safety data sources, including occurrence reporting systems and objective position data, showed evidence of mid-air collision risk between airplanes and helicopters near DCA in the years before the accident, but existing systems did not recognize or mitigate the risk.
We found that key stakeholders either did not have access to or were not making full use of the data.
For example, some of the systems used to analyze radar data for close proximity encounters in this investigation had rarely been used for that purpose before the accident.
We found that there was no standard definition to signify when aircraft are too close, which hindered the ability to track their incidents over time or to compare risk levels of different locations.
Lastly, although helicopters regularly triggered TCAST resolution advisories for airplanes approaching DCA, we found that helicopter operators were largely unaware of their involvement in these events and there was no standard mechanism to alert them after such events.
Staff have proposed two recommendations to the FAA in this area.
Next, excuse me.
Next, I will discuss the safety management systems that were in place at the FAA, PSA, and the Army at the time of this accident.
Multiple investigative groups coordinated to document and evaluate these systems.
The United States is a member of the International Civil Aviation Organization or IO, which sets standards for SMS.
It calls for inclusion of four main components.
safety policy, safety risk management, safety assurance, and safety promotion.
Building and sustaining a positive safety culture is also considered foundational to an effective SMS.
Both PSA and the FAA developed SMS programs using this model.
IO requirements do not apply to military aircraft operations.
The Army therefore had unique regulations and guidance for aviation safety programs.
In the remainder of my presentation, I will focus on SMS policies and implementation by the ATO and the Army.
At the time of the accident, the FAA had an established SMS for several of its organizations, including the ATO and ATO facilities such as the DCA tower.
Its SMS policy described the required four components of SMS.
However, this investigation revealed gaps in ATO's application of safety risk management and safety assurance in comparison to FAA's guidance for other service providers.
We found that unlike the FAA SMS guidance for operators which established requirements for coordinating with relevant external parties to collect and share safety hazard information.
The same requirement is not reflected in the ATO SMS guidance which is limited to bargaining unit representatives and management at FAA air traffic facilities.
There was no formal process at DCA for the FAA to share information about helicopter route traffic, TCAST resolution advisories, or potential conflicts with external stakeholders.
The formation of a helicopter working group was an example of an informal collaboration among multiple stakeholders designed to promote safety.
However, as previously discussed, that group was not able to affect changes such as relocating helicopter route 4 or adding hotspots to the route chart and spots with potential conflicts.
We found that the ATO's application of its SMS did not recognize and mitigate the risk of a mid-air collision at DCA, nor did it effectively coordinate SMS activities with external stakeholders.
As a result, staff have proposed two recommendations to the FAA in this area.
We also found that ATO management eroded the overall safety culture within ATO by not supporting its workforce, encouraging open communication, identifying and mitigating operational challenges faced by controllers, or fostering a just culture.
Interviews with ATO staff indicated that they feared retaliation for raising safety issues and management often discouraged filing reports because it made the facility look bad.
Interviews with current and former DCA tower personnel indicated that morale had been low for years before the accident due to the 2018 facility level downgrade and from the FAA's lack of transparency regarding the metrics used to support that decision.
External compliance verifications at DCA between 2022 and 2024 identified non-compliant items, including lack of staff support, poor communication between PTOAC Treycon and the DCA tower and concerns about potential conflicts with helicopter routes, but the ATO failed to recognize these items as indicative of problems at DCA.
Finally, DCA tower management personnel were reassigned after the accident, an action that appeared inconsistent with the characteristics of a positive safety culture.
As a result, staff has proposed one recommendation to the Department of Transportation in this area.
With respect to the US Army safety assurance policies, Army safety programs were required to incorporate an occupational hazard reporting and investigation program managed by unit aviation safety officers.
Aviation safety concerns could be submitted using operational hazard reports or OHRs, which required filling out and submitting a form detailing the issue.
OHRs submitted to safety officers were to be corrected at the lowest level possible.
When FAA facilities or operations were involved, it was sent to the nearest flight standards district office and to the Department of the Army regional representative.
The Army Safety Management Information System or ASMIS was another venue to report safety events including mishaps and near misses.
Mishaps and near misses were broadly defined and could include both aviation and non-avviation safety events and events that occurred on and off duty.
Although helicopters had triggered multiple airplane TCAST resolution advisories and they had been the subject of safety occurrence reports submitted by airplane pilots and controllers.
The Army's safety reporting programs did not provide information about those encounters.
An Army review of ASMIS found no reports of aviation mishaps or near misses near DCA area or OHRs referencing the DC helicopter routes or other hazards in the area.
Interviews with Army pilots suggested that participation in voluntary reporting programs was limited.
Additionally, the Army did not have a program to monitor operational flight data for its helicopters and as a result was unaware of routine altitude exceeded and related risks in the DCA terminal area.
Such a program could have not only helped identify altitude exceeded but may have also raised awareness about cumulative errors in the UH60 Lima's barometric altimeter system.
As a result of these findings, staff have proposed three recommendations to the US Army in this area.
In 2024, the Secretary of the Army issued a directive for the adoption of an Army Safety and Occupational Health Management System or ASOS, the Army's first official SMS framework.
The elements described in the directive are similar to IKEO SMS components described earlier.
The directive required incorporating ASOS in Army safety program regulations by September 2026 and it required Army commands and organizations to be fully compliant by the end of calendar year 2030.
Our reviews indicated that progress has been made in incorporating the directives requirements in policy documents, but ASOM's implementation has been slow, which may stem from several causes, including resource and staffing issues and a distribution of safety management responsibilities across Army organizations.
Additionally, we found that the Army lacked a positive safety culture in that its aviation safety system did not have the capacity, resources, or analytical mechanisms to consistently detect, interpret, and act on weak signals of latent hazards.
Overall, we found that the Army's process for allocating resources to aviation safety management did not ensure development of a robust SMS for helicopter operations in the Washington DC area.
As a result, staff have proposed two recommendations to the US Army in this area.
This concludes my presentation and we are now ready to answer your questions.
>> Thank you, Member Enman.
>> Thank you.
And I'll go back and Dr.
Price that was really helpful I think and for I guess I also want to because you kind of broke it down a little bit differently because I pulled some information out of our report.
I think it gets to the same place but we state that performance data analysis and reporting or PER indicates there were 4,067 encounters between 2018 and 25 in which separation between the helicopters and the fixedwing aircraft was less than or equal to 1,000 ft.
and 348 of those the separation was less than 500 feet.
To what extent did the FAA use utilize that data prior to this midair collision? Thank you for that question.
Um in the case of pars in particular um this after this accident was the first time that uh that system had been used in this fashion to look at the incidents and in fact it was it was a novel use of this system and a very useful one for our investigation but before this uh investigation pets had not been used in that fashion.
>> So I want to go back to the different ways that people can give information.
So the way I read it, there's currently the aviation safety action program or ASAP, the air traffic safety action program or ATP, the mandatory occurrence reports or moors, the near mid-agger collision systems, INMAX, the Aviation Safety Reporting System, ASARS, and the Army Safety Management Information System.
Now, collectively, it sounds like pilots and individual controllers were ringing a bell saying something's dangerous, do something.
And I'm just curious, is the FAA was they were they taking this information? Do they have the capacity? And will they take action on it? That's three questions in one.
I'm sorry, Dr.
Price.
>> Well, certainly there were numerous different reporting systems that provided information about near misses and events that were happening at DCA before this accident.
And the FAA did have access to all of those systems.
Um, our investigation revealed that in general, uh, the reviews of these systems were largely happening kind of on a case-bycase basis and weren't necessarily being looked at in an aggregate fashion.
And when we do a look back, of course, we can see that there was a definitive pattern in these data um, both with the safety reporting data and with the position data that I mentioned.
Um but as the uh FAA even noted in our public hearing um that was something that although they had opportunities to see they did not detect and mitigate before this accident.
>> Since that time have they indicated that they are using this data correctly.
>> Thank you for reminding me about that question.
Um since the accident we do have evidence that the FAA has put some of these data to use.
Um they did a a helicopter route analysis um in 10 different cities with uh 16 airports where they used um these data to look at and assess uh risk in in uh airports that had helicopter traffic and they have uh made uh some recommendations as a result of those.
Uh, one of the things that we remarked on though is that they are still not using a standardized approach for looking at that issue and it's something that we think is necessary to have more of a standard definition of what constitutes uh a close proximity event so that there can be a way of comparing one airport to another or looking at trends over time rather than um what is kind of a hodgepodge if you will right now of different ways of measuring this So I guess even though he didn't present captain just want to get your attention first are you under the impression whenever people file these reports somebody's actually listening >> I would say um generally you file the report and you don't hear anything back >> so what's the purpose of filing a report >> uh Dr.
price can probably answer this better, but I think goes into a database.
>> You're you're asking what is the purpose of a voluntary report? >> I guess I'm I'm I think a lot of people in aviation and a lot of us believe if you're saying something, somebody's looking at it or they're going to take action.
Is that a misperception? >> Well, no.
I I think that these safety reporting systems are vital and they serve an important purpose and we we would hope and we would like to believe that when a pilot or a controller files a voluntary safety report that it will be looked at and will be considered very seriously and like I said there is some evidence that um they there was some case byase analyses of reporting but I think the pattern was missed and That's where um we've made we've proposed recommendations that we think can get at that.
Thank you.
Um so we looked at uh for different um uh separations uh less than one nautical mile uh lateral, less than 400 ft vertical.
And this we this is something we worked out with FAA and looking at the data and that gave us the 15,214.
Then we looked at less than 600 feet lateral, less than 400 vertical, it gave us 210.
And then less than500 feet lateral and less than 200 feet vertical, which gave us 85, right? >> And so uh you know, so the FAA has 10 different sources of data that they're taking in.
It's a lot of data that they they request.
Uh they say they're data driven.
Atto did a top five every year and it says they're data driven.
I would say they are not.
Um, I was stunned, absolutely stunned when senators Jerry Moran and uh, Duckworth held a closed door briefing and invited me, invited the Army, invited FAA, and was asking us for an update and asking us questions and data came up and I talked about our 15,214 events that we've identified, the 210, the 85 five and FAA said actually it's five my five and you know this isn't the first time every time we release data that FAA gives us they come up with some new thing two hours later and it's just how you want to define it now my guess is so we had we and you tell me if I'm wrong.
You worked with MITER and the FAA on what the definition should be.
That was you and them working together to figure that out for our urgent safety recommendation.
Then right before right after we release it, they say that makes us look bad, I guess.
So they redefine it themselves and decide they should go with 600 feet lateral, 200 feet vertical to come up with five.
So you just so they is that correct? Well, Chairwoman Hammedi, I will be honest and say staff was never able to determine the source the exact source of where that five near misses quote came from in spite of requesting that explicitly.
Um, we have some ideas about what it could represent, but to be honest with you, we never got the clear clear response from FAA about the source of that uh five near misses quote.
>> Yeah.
Uh, so I I have to tell you and and we're going to we're going to get into this, but I um Dr.
price right before our hearing.
Actually, a maybe two months before a month and a half before the hearing, I got calls from staff saying, "We need your help.
I can't get information out of the FAA.
We need data.
We need help.
Can you help us?" I weighed in several times.
We finally got a response.
Then, uh, you, Dr.
Price, my my staff, and, uh, Mr.
Lauren Grath, uh, you all participated in a phone call with FAA.
FAA, you said, "This is the information we need." FAA said, "Okay, you know, put it in writing and we'll get it to you.
This is the agreement on what we'll provide." So, you'd put it in writing.
And the response was, "In in some cases, we were told that our requests were out of scope." Well, in this specific request, they basically told you no.
>> In this in this specific request, we were given some information about near midair collisions, but it didn't match the number five.
So, that was the source of the confusion there.
Um, but there was a response from Matt Kbeck that I have in writing that says after you they agreed to provide you information that they responded and said no.
I think I know what you're referring to and that was an instance where we had gotten some data from the ASAS program and we had additional follow-up questions about that and then as a result of that we were told that we needed to request that data directly from the air traffic organization and we followed up and did that and we were told that the air traffic organization would not duplicate work that had already done been done by ASAS >> even though part of the request test was an additional parameter that you gave them.
>> Yes, ma'am.
>> That was not a duplicate.
>> Correct.
>> Yeah.
Thank you.
Um, member Graham.
>> Thank you, Chair.
Well, I think uh many people would understand that the military has unique missions and and demands that necessitate deviations from what might be standard or accepted in the commercial world.
A proper comprehensive safety management system should not be one of those things.
As well documented by our team in their presentation, neither the Army's safety management information system nor their safety and occupational health management system, individually or combined, comprise a an acceptable SMS.
I have relentlessly beat the drum of SMS across multiple modes of transportation and organization since I arrived here at the NTSB.
And uh I truly believe that the military should be no different on this one.
The core pillars of SMS are intentionally broad concepts that can be adapted to any operation.
In particular, any operator regularly conducting flights in class B or congest other congested airspace should have an SMS to properly manage their risk.
Our airspace is only as safe as its weakest aircraft or least safe aircraft or organization.
Dr.
Price, uh, I got a question or two for you and maybe somebody else might want to jump in on this too.
I don't know.
As you mentioned, the Army Safety and Occupational Health Management Systems included some elements of a similar was what's similar to an IKO SMS components.
While that is true, is it also true that this program was not a standalone for the Army Aviation Battalions? >> Member Graham, I'd like to ask if Dr.
Bramble would address that question.
>> Yes, you're correct.
Um, aviation was just one application of many of their safety and occupational health management system.
>> Okay.
Um, what other categories of potential hazards or other than aviation safety were included in this program? Uh there was a wide variety including tactical, offduty, automotive, um u many things unrelated to aviation.
There were 14 different areas I believe.
>> Well, I'm I'm not going to say that those aren't important.
They are um they are important for each unit.
But I think what became abundantly clear during the course of this investigation is that the army aviation safety army aviation safety cannot simply be lumped in with all of those other elements.
Is it true that the Army's safety and occupational health advisory council which met bianually to review among other things workplace safety hazard logs, work-related illness and injuries, civilian case incidents, worker compensation costs and installation hazard abatement? >> Yes.
And while these bianual meetings could also cover the items cover items directly related to aviation safety, can you share why this does not meet the level of what you would expect to see through like an aviation safety action program or a FOA program or in the granddaddy of them all the ASA system? >> Sure.
I mean uh I guess the first difference would be that it's integrated with the command structure and run by the commander and in the uh in the case of an event review committee or something like that you'd have sort of the safety and QA folks working it independently and advising management.
Secondly, those types of committees in commercial aviation tend to have much better data streams um such as FOQUA data, LOSA, ASAP, that sort of thing.
Yeah, they they had no real they had an employee reporting system.
Is that correct? >> Yes, they had uh two different ways that uh reports could be submitted.
>> And were there many reports to look at? >> No.
In the case of occupational hazard reports, there hadn't been any received by the brigade in the last year.
And in terms of the aviation safety management information system near miss reports, there had been five from the 12th aviation battalion, three wildlife strikes, one case for a crankshaft left open on the ground, and one offduty report.
What benefits do you believe the uh army would gain from having a standalone fully resourced SMS program dedicated to aviation safety, which they don't have today? Well, they would likely have better resources, uh, better focused aviation focused analytic capabilities, better data streams, and a better focus on identifying weak signals of latent risks in the system that represent low probability, high consequence accidents, the kinds of complex system failures we're talking about today.
>> Great.
Thank you, Dr.
Bramble.
>> Member Emman.
So I want to go to the uh actually in the safety data group chairs analysis report uh we indicated that the aviation risk identification and assessment or the ARA system was designed specifically to identify air traffic operations that represent potential safety risks even if operations are technically deemed compliant.
We received testimony I think from the FAA eastern or safety group for the eastern service area that Arya looks at the interaction of aircraft.
So how they are moving towards each other and generates a report.
I think we looked at I believe 874 of those just between 2022 and 24.
Um to what extent did the ATO use this risk and this data to understand prior to the midair collision what was going on? >> Thank you.
Um the ATO was using Arya as you specified.
Uh they were reviewing on a casebyase these 874 events that they referred to as preliminary Arya reports.
The system itself would generate something for them to review.
When aircraft came within uh a certain proximity, an algorithm would feed a report to them for review.
They had quality assurance teams that would conduct those reviews and assess and what we found was that their assessments were largely through the lens of compliance.
So, uh they were looking to see if there was any kind of lack of compliance and if there and if there was they would they would categorize it and if if not they would they would categorize it into another category.
Um, interestingly, within that set of 874, none of them had been identified as being near mid-air collisions, even though we knew from other sources during that time period that pilots had made those reports.
So, that was what made us concerned and feel that there needed to be other ways of looking at these data in a systematic way.
And we've made uh proposed recommendations that we think can address that.
It's interesting you use that word.
Is is in your opinion or just what you viewed is the FAA more interested in enforcement or safety? >> Well, in with respect to these data, I do think that the ATO in particular was looking at some of these data through more of the compliance lens.
Um, the FAA certainly has the capability to look at data in a more holistic way.
We uh do have evidence that for example the ASAS program is doing that and that it is being done.
I don't want to say that it's not being done but in this instance with these particular types of data we did feel like they weren't looking at it in in kind of a an aggregate or big picture risk focused but rather a compliance-based way.
>> And I think that's a nice way of saying it probably but I I would offer again the same thing.
Are we going to five years from now be saying we're seeing something that the data was there again? Because it seems like every five every two or three years it's almost whack-a-ole.
We continually go back and say, "Oh, they're looking at it.
They're trying." And then we just come back and we have another incident.
Well, one of the recommendations that staff has proposed has to do with establishing asking FA to establish a standard measure of close proximity, making that public and then creating a database so that all of the stakeholders can have access to those data so that they so that people can conduct analyses and be aware of these types of events.
>> So, is it the public's responsibility or is it FAA's? Well, we would certainly expect FAA to analyze those data, but because those data are available and can be public.
There are a lot of different stakeholders who who are interested in this and having that data available to a wide range of people is one way to ensure that it can be uh there can be transparency and can be looked at.
>> I guess let me ask I guess a different way.
I'm kind of curious because we can go through this report and a lot of different things.
I'm just if you could change one thing at the FAA today and make everyone safer, what would you do? I'm not trying to lead you into a question, Dr.
Price.
I'm just interested in your opinion.
>> Well, sir, I think everyone at this table would have an idea of of what they what they would want to do.
I am I am a safety data person.
So um you know I would like to see uh better sharing of safety data information, better analysis of safety data information and the the recommendations that that our group has proposed in this area we hope can can move forward in that because then it can allow for better foresight and better prediction of risk in the system.
>> Okay.
Thank you.
>> Thank you.
Um I wanted to talk about sharing of data uh which you mentioned with external stakeholders.
Um can you talk about that a little bit first? How does NTSB get its data? >> Uh Chairwoman Tomy, there's a variety of different ways we get data.
Um we we we we own some data.
We are the shepherd or the the stewards of some some aviation data.
Um, we access publicly available data.
We can request from FAA lots of different kinds of data.
Um, and there are some sources of data that are proprietary.
For example, um as as I mentioned, the ASAP and ASAP voluntary reporting systems and the FOCA systems that are part of the ASIS program that are are uh proprietary and we have a special agreement memorandum of understanding with FAA and there are certain um situations where we may uh request those data.
>> Some data systems that FAA have we still pay for.
>> That is correct.
That is correct.
>> Instead of the sharing of data.
So how about industry? How does industry get say the commercial airlines? How would PSA get information from FAA? >> Well, PSA as are many airlines are members of the ASA program which is a collaboration between industry um and and uh well it's a it's a collaboration among many stakeholder groups that's managed by uh a third party um that pro that brings information together and and allows them access to it.
So, that's one way that PSA and many uh operators get safety data.
>> Some of the criticism of that though is that it's outdated and old by the time they get it.
>> Well, we were told in in this investigation that there were lags between when they were putting uh contributing data and when they could see that data.
And so um >> how long? >> Uh sometimes three months.
>> How about the army? Do they get the data? >> We asked Army about their use of FAA data.
Um before the accident, they did not routinely request uh data from FAA.
um in the in the public hearing they said that they were interested in starting that practice and we uh queried them again recently and we do know that they have um made some outreach efforts to FAA.
We know that there has been some conversation about them becoming participants in the ASIS program.
Um, I think there has been there have been some challenges with that and again I think some of the uh recommendations that staff is proposing we hope will help to foster that type of sharing.
>> Did the FAA ever give the Army data? >> To the best of my knowledge, no, they did not.
>> Right.
Um, and we'll get to Well, what are the biggest barriers I guess to to that information sharing? >> Well, I think that some of the barriers that we identified was that there wasn't a common language of sharing some of this information.
So, there wasn't kind of a common common language.
We also just found that some parties, as we mentioned, as we just alluded to, didn't have access to some of the data sources.
Um, and and in some cases, weren't even aware of it.
And I think another barrier that we saw was that even though there were this there was this kind of ongoing stream of reports coming in that those individual reports those NEAC reports those ASAP ASAP reports weren't leading to the questions and looking at things in a bigger scale looking at kind of a is this a system approach and I feel like that linkage was also kind of missing.
Okay, one more question on the on the data side.
I might Can I ask Mr.
Graph a question? >> Certainly.
>> Hey, Lauren.
Uh you and I have talked about this that there are are some concerns uh with uh an over reliance on AI by FAA that there are some challenges there that if there uh that they've got to be careful on the use of AI to pick um up trends to make sure it doesn't discount uh some reports.
Can you just talk about some potential um benefits you could get from the use of AI to help them identify trends but then also some some potential downsides? >> Sure.
Thank you for that.
Um it was even there was some discussion even in the investigative hearing on use of AI.
The the term is widely used to include a lot of uh kind of algorithmic computer process.
Um and so what is it that you're actually talking about when you when someone says AI as a question and I think um in this case it is being used FAA is using it uh in some cases when they have large volumes of uh pilot reports to to make sense of them categorize them and group them and it directs attention there.
I think one of the things um even the Arya system could be considered to be AI.
I think what we saw here was the in some cases was not really understanding what was represented.
So there really does need to be a human understanding of what all of these things mean together.
So if you're if you're looking to a computer system to give you the answer and kick out the answer that I think that would be problematic if it would be that type of a an application.
it really does need a human understanding and review uh to to trust what they're getting out of it.
And it so it may be that if there's any um loss of true understanding of what those events mean, that could be problematic.
So, pluses and minuses.
>> Yep.
Thank you, uh member Graham.
>> Thank you, Chair.
I'm just amazed about the amount of data the FAA air traffic control organization had and it sure doesn't seem like they were doing anything with it.
Was the FAA aware or had they been aware made aware of the risk of a mid-air collision near DCA along Route 4 and the approach path to runway 33 prior to the accident? >> U member Graham, I I think the answer to that is yes.
There were safety reports and they there was also the work of the helicopter working group that had raised um these issues as a concern.
>> Yeah, I agree.
It was there.
I think it was there.
People were telling them it was there.
One of the things that struck me as I read through the multitude of of uh data uh the FAA had available to them was now nearly every subjective data reporting mechanism was inconsistent with the objective data source events surrounding DCA.
For example, I want to draw uh attention to the discrepancy between ATSAP program for air traffic controllers and the ASAP program for pilots.
Over a nearly 4-year period from February 2020 through October 2024, more than 16,000 ASAP ASAP reports were filled uh or were filed related to DCA.
That's from the the pilots, including 85 specifically concerning close calls between airplanes and helicopters.
Yet somehow over nearly a 12-year period, three times as long from January 2011 through August 2023, only 520 ADSAP reports were filed related to DCA.
Of those, just 26 contain information on close calls between airplanes and helicopters.
To put those numbers in perspective, pilots at DCA were filing nearly 18 ASAP reports annually regarding close calls between airplanes and helicopters.
And controllers at a DCA were only filing two per year.
I'm going to ask Mr.
Soer this.
With regard to the discrepancy between ASAP and ADSAP programs uncovered during this investigation, what do you believe is the reason for air traffic controllers identifying significantly fewer close calls than the pilots? Thank you for the question.
So in air traffic control there is a little bit of a difference.
So that that metric could be a little deceiving.
And that's because in air traffic we also have you've heard mentioned a couple of times mandatory occurrence reports which are requirements for air traffic control to submit uh you know losses of separation would be one of those things.
Near midair collisions would be one of those things.
Runway incursions would be one of those things.
There's the list goes on.
So we looked at uh I believe it was 10 years and Janna please correct me if I'm wrong like 10 years worth of data approximately and they had 170,000 plus mandatory occurrence reports at DCA that were reported by air traffic control.
So for air traffic controllers, they may not file an ASAP based on a report, especially if they've already reported it via mandatory occurrence report because they feel they've met a a reporting requirement and they didn't go they don't take the time to go actually do an ATSAP on top of it.
>> Is is it difficult to do an ASAP report? >> No, I would say it's not difficult to do one.
They may >> Is there a reluctance? >> Um, no.
was well from our investigation here at DCA, none of them displayed a reluctance or or um said that they had a reluctance.
They felt free to file and they did.
They did file reports.
>> Was there any fear of retaliation for reporting in any way? Uh close calls, >> not according to our interviews.
They didn't feel a fear of retaliation for reporting in an ASAP.
>> Okay.
Thank you for that.
Um, I think that's all I have right now.
I'm gonna wait for the next round.
>> Member Emman, >> I'm actually I I think I'm done with my questions and so I'll either Can I pass until we get to the last round? >> Cool.
>> Okay.
Because I do have a couple of last minute things.
>> Dr.
Bramble, Dr.
James Reasonz talks about five components of a positive safety culture.
Reporting culture, just culture, flexible culture, learning culture, informed culture.
Do you believe the FAA had a reporting culture? And let me just for everybody listen watching a reporting culture is a culture where staff feels this is per James Dr.
James reason feel safe and confident reporting safety concerns near misses and errors without fear of blame.
>> I should probably defer this question to Dr.
Wilson.
So in some regards we believe that the facility at the facility level there was a reporting culture.
Um controllers did indicate that they felt comfortable submitting ASAP reports and management also said that they had an open door policy and controllers said that they felt comfortable going to their managers if they had a safety concern.
At a more global level at the ATO we do not believe that there was a reporting culture.
Um, as was mentioned in Dr.
Price's presentation, um, some staff did feel that there, uh, would be retaliation for reporting safety concerns.
We had multiple, um, people who wanted to share information with us, but didn't feel comfortable until they were closer to retirement because they were afraid of retaliation.
Um, we obviously had the issue at the public hearing, our investigative hearing, where we felt like management might have been influencing the testimony of some of our witnesses.
So um in some aspects yes and some aspects no.
>> In fact we couldn't get pe some people absolutely refused to testify or even be interviewed on the record.
Uh one person in particular went to someone a friend of mine at the FAA to call me to ask if he could meet with me because he wasn't comfortable talking to anybody else.
It's inappropriate for me because I'm not an investigator.
But I was able to connect uh through that intermediary.
I don't even know the person's name who came forward, but uh with with Mr.
Soer and Miss Lewis because they were afraid.
People who testified they were afraid.
I mean uh and I'll ask Mr.
Sober if you can comment on that.
I mean numerous times people indicated they were afraid of retaliation.
We did have several.
Now, just wanted to differentiate, not people that were actually at the facility in this case, but yes, we had uh I had at least three that I felt were key concerned personnel that absolutely refused, >> including all our witnesses that were concerned.
>> Yes, the witnesses that we did get were very concerned.
>> Uh how about the reporting culture at the army? It was not robustly implemented at the unit level.
>> How about just culture? Just culture and atto.
Just culture means an atmosphere of trust where people are encouraged to report essential safety related information, but clear lines are drawn between acceptable and unacceptable behavior.
Again, I think at the facility level um for controllers, they were encouraged to report ASAP reports, but um we do know that following this event, um managers were reassigned.
Um as Dr.
Price mentioned in her presentation, um following some mandatory um report uh occurrences that occurred, um they were told not to report them because it might make the facility look bad.
Mhm.
>> And then again just those hesitant to talk to us.
I think uh goes to show that there was not the just culture.
>> Just culture at the army.
>> We did not find that to be a problem at the unit level in the army.
>> Flexible culture.
Do you mind if I finish the five? Yeah.
Flexible culture.
Culture that can adapt effectively to changing demands and when necessary shift from normal hierarchies to a structure that relies on expertise.
atto.
>> I think we found that the ATO did not adapt to uh recommendations concerns over near midair collisions in the DCA area.
>> Flexible culture army.
>> I think that um the safety assurance system was not nimble enough to specifically look at some of the risks in the DC airspace.
Informed Culture, an organization that collects and analyzes relevant data and actively disseminates safety information to the workforce.
ATO >> again as doc as Dr.
Price mentioned with the safety assurance process, we definitely found failures with that.
Um atto failing to share information with parties that were involved in some of these occurrences and also on a broader sense failing to identify the midair collision risk.
>> Army Uh this was definitely a weakness.
Uh they lacked information about the frequency of near midair collisions between their helicopters and other aircraft in the airspace and they were unaware of the prevalence of maximum altitude exceeded among their helicopters >> which which and they could have gotten through flight data monitoring.
>> Correct.
>> Uh informed culture an organization that collects an an oh wait did I do that one? learning culture.
An organization capable of learning from its mistakes, analyzing data, and making necessary improvements, ensuring staff understand SMS ATL.
Again, um numerous times the helicopter working group brought recommendations to the ATO which were not acted upon um and those were ignored.
We also had external compliance verification findings over multiple years that the um the ATO did not learn from >> Army.
>> I think they might have if they had been aware, but I think because they were not aware of these risks, they did not learn about them.
>> Thank you, member Graham.
>> Thank you.
I want to talk a little bit about the FAA SMS.
Um just amazing.
Uh I think we found that there were some serious issues with their SMS and uh we'll talk about those now but it it just it's the height of hypocrisy when a regulator expects operators it regulates to follow prescribed uh described SMS regulations to the detail but can't seem to come even remotely close to following its own in this case their own SMS.
Um let's start with safety risk risk management which is the second pillar of an SMS.
Uh what did the SRM safety risk management entail at ATO? Uh the ATO did have safety risk management um policies in place that we did find on several occasions.
It was followed particularly at the facility um following the GAO noise study.
There was a SRMP that was conducted in March of 2023 which revised some of the altitudes along the helicopter routes.
Um however we did also find that um SRM wasn't typically used uh in response to issues identified through um that were identified through the safety assurance process.
>> So they weren't doing them very often, were they? >> It appeared to be dependent upon the issue.
>> Hit or miss.
Hit or miss.
Yeah.
I've been part of one of those processes in my prior life.
So um and let's talk about safety assurance.
When they uh when safety insurance activities identify deficiencies, what actions should be taken according to the ATO's SMS might I could take that.
>> Thank you.
>> Safety risk management and safety assurance are are considered uh sort of complimentary activities.
So safety assurance is really just monitoring uh safety performance indicators to ensure that the safety risk management is working as intended.
So it would it would be sort of a feedback then if the safety assurance identifies uh performance that doesn't match the SRM that it would it would feed back into the safety risk management process to re-evaluate whether there was maybe hazards or risks that weren't properly assessed or controlled.
didn't seem like that was happening very often.
Yeah.
Finally, let's talk about the fourth pillar of SMS.
And I could talk about all the pillars, but we don't have enough time today.
Um, safety promotion.
What was that supposed to entail at the ATO? So safety promotion would be um focused on SMS training for personnel um disseminating safety lessons learned and these actions would all uh foster a safety culture at the ATO.
>> We definitely saw some issues there, didn't we? >> We did.
We learned in the hearing that um Mr.
Fuller stated that there was no SMS training for ATO personnel.
Um it was the expectation that management at the facility would have access to the manual and therefore be familiar with SMS because of that.
>> That's that's at the foundation.
Every employee is the safety officer.
They need to be trained in what their responsibilities are and their roles are.
I don't get it.
I know this is asking your opinion, but objectively in our review of ATO's safety culture, did we find any evidence that it was a positive safety culture without actual or fear of retribution for employees who voluntarily reported safety concern concerns? We don't have evidence that anybody who reported uh a safety concern via ATSAP faced any retaliation.
Um but again, we did have uh several people who talked to us who said that they faced retaliation for raising safety concerns at the ATO.
So would it be fair to say that their SMS was nice on paper or but not maybe not so nice in in reality? >> They did have an SMS um manual.
So they did have it documented.
However, we did identify a number of deficiencies and that is why we have a recommendation for their SMS to be evaluated.
And in your judgment, had the ATO followed their prescribed SMS with regard to what they knew about the risk around DCA before this accident, would that process led them to enact mitigations that could have prevented this accident? >> We'd like to believe that a properly implemented SMS would have identified this.
>> I agree with that.
Thank you.
>> Good questions.
Member Emman, >> I guess this may be our last round.
Okay.
So, um, first off, the conversation was good.
I appreciate it.
Member Graham, chairman.
It reminds me of something actually of someone who just passed away.
It was a great comment.
When culture clashes with policy, culture always wins.
That was Fred Smith at FedEx.
I think it's a very apt quote.
So, I I've done something every board meeting um since I've been on the board.
So, I'm actually going to ask Deputy Managing Director Brian Curtis.
I think it's important.
Can you tell us how many current investigations we have ongoing are open at the NTSB? >> Yes.
Currently, across aggregate of all the modal uh offices, there are 87 open investigations.
>> Okay.
And uh Mr.
Learon, how many do we have open in aviation alone? So, presently we're at about 711.
Um, and to put that in perspective, we do about 1,200 a year.
Yeah, I only bring that up in that uh it's probably be the last time that I'll have a little bit of opportunity to to before chair do a closing statement, but um I know this is tough on everyone, but there's a lot of other people that these investigators have to work and try to help.
And I know whenever I was watching the board meetings, I started realizing quickly not everyone gets the full board meeting, but there's still a lot of important work that's being done by our investigators and a lot of people and especially victims families that are out there hoping for answers.
So, I'm I'm I'm appreciative to the staff for that.
It just helps us to remind it a little bit.
Uh and it also is it's a reminder for me.
It's it was I launched on 91 aviation fatalities last year.
I did 13 family briefings and I am tired of doing them and I'm sorry for you because these the pages of these reports are written in your family members blood.
So with that again I am sorry that we have to be here.
I know we'll get into the final final calls and a lot of things shortly but uh hope this is helpful in what we've been doing and what we will do after tomorrow.
Thank you.
>> Thank you.
Well said.
Uh, I do have um a a question though, Mr.
McKenna and Dr.
Bramble, maybe you as well.
How would you describe uh your work and the willingness your work with the army throughout the investigation and their responsiveness to NTSB requests? >> Uh, generally the cooperation was excellent.
Um we received some delays later in the process.
Um but by and large they were pretty responsive.
The army was uh excellent and responsive to work with especially through the uh brigade.
Um we had no problems getting the information we requested and they uh rapidly provided us assets such as aircraft to do flight tests.
So um it was good experience, good uh working relationship with the army at the tab level.
M >> Mr.
Banning, would you agree? >> Yes, absolutely.
>> Yeah, I see uh Colonel Easter and uh Colonel Jagger, I believe, uh in the um audience and then we also worked with Colonel Parker.
I thought it was an excellent uh working relationship.
I they provided us what we needed.
I do want to thank Secretary Driscoll.
We speak often.
He was very responsive when I uh we needed assistance and I really appreciate that.
Uh Mr.
Soer, maybe others who might want to Well, let's start with Bryce.
Mr.
Banning, how would you describe your work with the FAA throughout the investigation and their responsiveness to NTSB requests made of the ATO? I think we are grateful for the FAA's role.
We're grateful they did provide us a tremendous amount of data.
Um we're grateful for their cooperation.
Um however, there were some struggles that have already been stated.
Um it is difficult when conducting an investigation to be told what you what is within the scope of your investigation.
So that >> telling us what is and out of scope.
Yeah, that was problematic.
Um, but we do appreciate their role and they did provide us a tremendous amount of data.
>> Absolutely, we appreciate their role.
However, there were a lot of data and in information challenges and concerns on retaliation and everything else.
I will say I think it's very interesting that for all the data we requested and much of what what we did not receive you received a phone call on Friday or email and phone emails from FAA asking two different emails I believe did we have everything we needed and did we get all of our data requests answered on Friday.
Friday.
That's ridiculous.
One year later, six months of asking, actually, I think you asked me, you all asked me to start weighing in back in May.
Some of those questions never answered.
And yet Friday they ask, "You're too late.
We already figured it out ourselves." Okay.
Well, any other questions by Nope.
All right.
Well, we're going to take a uh 15 minute 15 minute break.
Uh we'll be back uh using 15 minutes.
Yeah.
Four o'clock.
We are back in session.
Mr.
Curtis, will you please read the proposed findings? >> Certainly.
Chair, as a result of the investigation, staff proposes the following 71 new findings.
Number one, the pilots of flight 5342 were certificated and qualified in accordance with federal reg regulations.
Number two, the pilots of flight 5342 were medically qualified for duty and available evidence does not indicate that they were impaired by effects of medical conditions or substances at the time of the accident.
Number three, review of the flight 5342 pilots time since waking and sleep opportunities in the days before the accident indicated that the pilots were unlikely to have been experiencing fatigue.
Number four, the pilot, instructor pilot, and crew chief on board PAT 25 were qualified and current in their positions as designated by the unit commander in accordance with Army regulations.
Number five, the pilot, instructor, pilot, and crew chief of PAT 25 were medically qualified for duty and available evidence does not indicate that they were impaired by effects of medical conditions or substances at the time of the accident.
Number six, review of the three PAT 25 crew members time since waking and sleep opportunities in the days before the accident indicated that the crew were unlikely to have been experiencing fatigue.
Number seven, the airplane was properly certificated, equipped, and maintained in accordance with 14 CFR part 121.
the the airplane operated within its weight and balance limitations throughout the flight.
Examination of the >> Sorry, can you can you pause for a second? I'm sorry.
Yeah.
Okay, we're back to seven.
It had flipped to eight.
Go ahead.
>> You're fine.
Start at seven.
It just the video the screen was off.
>> Okay.
I'm sorry.
Number seven.
The airplane was properly certificated, equipped, and maintained in accordance with 14 CFR part 121.
The airplane was operated within its weight and balance limitations throughout the flight.
Examination of the airplane revealed damage consistent with an in-flight collision and subsequent impact with water, and there was no evidence of any structural system or power plant failures or anomalies.
Review of surveillance videos indicated that the airplane's wing navigation, landing taxi, and anti-corrosion anti-colision strobe lights were operating at the time of the collision.
Number eight, the helicopter was properly certificated, equipped, and maintained in accordance with US Army regulations.
Review of helicopter maintenance records did not reveal any open discrepancies or anomalous trends that contributed to the accident.
The helicopter was operated within its weight and balance limitations throughout the flight.
Examination of the helicopter revealed damage consistent with an in-flight collision and subsequent water impact.
There was no indication of any structural main or tail rotor system, flight control system, or power plant failures or anomalies.
Review of surveillance videos indicated that the helicopter's right and tail position lights, the landing light, as well as both upper and lower anti-colision lights were operating at the time of the collision.
Number nine, the operations supervisor and four controllers who were working in the Ronald Reagan Washington National Airport air traffic control tower cab at the time of the accident were properly certified qualified in accordance with federal regulations and facility directives and current.
Number 10.
Although the Ronald Reagan Washington National Airport air traffic control tower facility was not staffed to its large target level at the time of the accident, the number of staff in the tower at the time of the accident was adequate and in accordance with Federal Aviation Administration directives.
Number 11.
The decision to combine the helicopter control and local control positions was not the result of insufficient staffing and personnel were available to staff the helicopter control and local control positions separately had the operations supervisor chosen to do so.
Number 12.
The local control controller, assistant local controller, and operations supervisor were medically qualified for duty, and available evidence does not indicate that they were impaired by effects of medical conditions at the time of the accident.
Number 13, review of the local control and assistant local control controllers and operation supervisors time since waking and sleep opportunities in the days before the accident indicated that the controllers including the OS were unlikely to have been experiencing fatigue.
Number 14.
Visual meteorolog meteorological conditions prevailed in the area at the time of the accident.
A review of observations recorded throughout the night of the accident revealed no evidence of any local atmospheric pressure anomalies that would have impacted barometric altimeter readings.
Number 15.
Keeping the helicopter control and local control positions continuously combined on the night of the accident increased the local controllers workload and negatively impacted his performance and situation awareness.
Number 16.
Had the helicopter and local control positions been staffed separately, PAT 25 might have received a more timely and effective traffic advisory.
Number 17.
Due to extended time on position at the time of the collision, the operations supervisor was likely experiencing reduced alertness and vigilance, which decreased his awareness of the operational environment and reduced his ability to proactively assess the risks posed by the traffic and environmental conditions at the time of the accident.
Number 18.
The lack of mandatory relief periods for supervisory air traffic control personnel is contrary to human factors research that shows clear performance deterioration in situations of prolonged time on task.
Number 19.
Although the local control controller provided an initial traffic advisory to the crew of PAT 25 in accordance with Federal Aviation Administration job order job order 7110.65.
He did not provide a corresponding advisory to the crew of flight 5342 regarding Pat 25's location and intention which could have increased situation awareness for the crew of flight 5342.
Number 20.
If the LC controller had issued a standard safety alert to the flight crews of either aircraft as prescribed in FAA order job order 7110.65 providing the conflicting aircraft's position and positive control instructions, they could have taken immediate action to avert the impending collision.
Number 21.
Initial and recurrent scenario-based training in threat and error management would help controllers identify and mitigate risks and strengthen awareness.
Sorry, strengthen situational awareness situation.
I'll read that again.
21.
Initial and recurrent scenario-based training in threat and error management would help controllers identify and mitigate risks and strengthen situation awareness.
Number 22.
A risk assessment or decision-making tool would likely have benefited the accident OS in identifying and mitigating the operational risk factors that were present on the night of the accident.
Number 23.
Due to degraded radio reception, the crew of PAT 25 did not receive salient information regarding flight 5342's circling approach to runway 33.
Number 24.
The PAT 25 IP did not positively identify flight 5342 at the time of the initial traffic advisory despite his statement that he had the traffic in sight and his request for visual separation.
Number 25.
With several other targets located directly in front of the helicopter represented by points of light with no other no other features by which to identify aircraft type and without additional position information from the controller, the IP likely identified the wrong target.
Number 26.
interference that obscured the controller's circling to call the microphone keying that blocked the PAT 25 crew from receiving the instruction to pass behind ambiguous ambiguous visual clue cues and the lack of an integrated traffic awareness and alerting system likely reinforced the PAT 25 crews expectation bias that the airplane was among the traffic approaching runway one and did not pose a conflict.
Number 27.
The absence of documented training on Ronald Reagan Washington National Airport's fixed wings procedures and the mixed traffic operating environment represented a safety vulnerability for Army flight crews operating in the Ronald Reagan Washington National Airport Class B airspace.
Number 28.
Due to additive allowable tolerances of the helicopter's pedo static altimeter system, it is likely that the crew of PAT 25 observed a barometric altimeter altitude about 100 ft lower than the helicopter's true altitude, resulting in the crew erroneously believing that they were under the published maximum altitude for route 4.
Number 29.
A recurrent task to verify the continued accuracy of recorded flight data for US Army aircraft would help ensure the data integrity needed to support quality assurance and safety programs and accident investigations.
Number 30.
The Federal Aviation Administration and the Army failed to identify the incompatibility between the helicopter routes low maximum altitudes and the error tolerances of barometric al alimeters which contributed to helicopters regularly flying higher than published maximum altitudes and potentially crossing into the runway 33 glide path.
31.
Pilots need all available information on their the potential total error allowed by design that could occur in flight on any on an airworthy barometric altimeter.
Number 32.
The Army's postinstallation functional check of the transponder on the accident helicopter was insufficient to detect that it was not broadcasting automatic dependent surveillance broadcast out.
Number 33.
The Army's lack of a recurrent transponder inspection procedure resulted in the incorrect aircraft address being transmitted by the accident helicopters transponder and the incorrect automatic dependent ser surveillance broadcast settings on several other helicopters being undetected or 34 because the APX-123A transponder is designed for use on multiple aircraft platforms.
It is possible that incorrect settings may be present on other aircraft used throughout the Department of War Arms Services.
Number 35.
The crew of Flight 5342 did not see the helicopter until it was too late to avoid a collision because of the high workload imposed during the final phase of their approach and due to the helicopter's low conspicuity and lack of apparent motion.
Number 36.
Times of compacted demand as a result of air carrier scheduling practices increased operational complexity and required mitigation by controllers to maintain spacing and surface movement.
Number 37.
Ronald Reagan Washington National Airport air traffic control tower routinely received less than the requested miles and trail spacing from PTOAC consolidated terminal radar approach control which increased controller workload by requiring them to generate additional spacing to prevent delays or gridlock.
Number 38.
the practice of offloading arrival arrival traffic on approaching on approach to runway one by asking pilots if they could accept a circling approach to runway 33 was a routine mitigation strategy for Ronald Reagan Washington National Airport controllers to generate spacing that was not provided by PTOAC consolidated terminal radar approach control 39 timebased flow management or metering would provide PTOAC consolidated terminal radar approach control and Ronald Reagan Washington National Airport air traffic control tower with a consistent flow of traffic with more accurate spacing and greater predict predictability thereby reducing controller workload.
Number 40.
Ronald Reagan Washington National Airport air traffic control tower has significant airspace, airfield, mixed fleet, and operations complexities that appear to be inconsistent with its current facility level classification.
Number 41.
The Federal Aviation Administration Air Traffic Organization failed to recognize external compliance verification results as indicators of systemic traffic management volume and flow issues at Ronald Reagan Washington National Airport for which controllers were required to compensate.
Number 42.
the long-standing practice of relying on pilot applied visual separation seen avoid as the principal means of separating helicopter and fixedwing traffic in the Washington DC area by both Ronald Reagan Washington National Airport air traffic control tower and the army led to a drift in operating practices among controllers and helicopter crews that increased the likelihood of a mid-air collision.
Number 43.
Reliance on pilot applied visual separation seen avoid as a primary means of separating mixed traffic introduced unacceptable risk to the Ronald Reagan Washington National Airport Class B airspace.
Number 44.
Ronald Reagan Washington National Airport air traffic control towers procedure of maintaining maintaining a discrete helicopter frequency when the local and helicopter control positions were combined decreased the overall decreased overall situation awareness for pilots operating in the area.
of 45.
Providing controllers with additional salient cl cues regarding the perceived severity of a potential conflict would reduce controller cognitive load and would likely improve reaction time to the most critical conflict alerts.
Number 46.
There was no evidence that the local control controller, assistant local control controller, or operations supervisor were under the influence of alcohol or prohibited drugs at the time of the accident.
However, evidence was substantially limited by the lack of post-acc alcohol testing, and evidence was of somewhat lower quality that it would have been if drug testing had been conducted sooner following the accident.
Number 47.
The Federal Aviation Administration Air Traffic Organizations ATO drug and alcohol testing determination did not meet Department of Transportation DOT timeliness requirements.
Furthermore, the ATO's decision to not conduct drug testing as soon as possible after the testing determination and to not conduct alcohol testing at all violated DOT requirements.
Number 48.
The delayed and in inappropriate drug and alcohol testing determination was due in part to the air traffic organizations ATO determination process being inadequately designed to routinely meet Department of Transportation requirements for timely testing and in part to ATO's staff incomplete understanding of those requirements.
Number 49.
Additional reviews of helicopter route charts as required by Federal Aviation Administration order 7210-3D would have provided an opportunity to identify the risks the risk posed by the proximity of Route 4 to the runway 33 approach path.
But there is no evidence to support that these reviews were being performed at Ronald Reagan Washington National Airport.
Number 50.
The information published by the Federal Aviation Administration regarding Washington DC area helicopter roads was insufficient to provide helicopter and fixedwing operators with a complete understanding of the helicopter route structure and its lack of procedural separation from fixedwing traffic.
Number 51.
Current aeronautical charting does not provide information on visual flight rules, helicopter routes that may conflict or come in close proximity to and approach departure corridors which reduces pilot situation awareness.
Number 52.
The lack of automatic dependent surveillance broadcast ADSB out from the accident helicopter did not contribute to the act this accident as the helicopter was still being tracked by radar.
and ADSB out would not have provided improved traffic alerting from the Ronald Reagan Washington National Airport controller or the crew of flight 5342 because the airplane was not equipped with ADSB in number 53.
The Army's Army's standard operating procedures that prevent flight crews from enabling automatic dependent surveillance broadcast ADSB out while in flight when not performing sensitive missions that require ADSB to be disabled.
Limit the visibility of military aircraft on collision avoidance technologies that leverage ADSB information.
Number 54.
Although the airplane's traffic alert and collision avoidance system operated as designed, it was ineffective in preventing the collision because of current activation criteria and resolution advisory inhibit altitudes.
Number 55.
Traffic advisory oral alerts that include additional information about the location of traffic could reduce the time pilots need to visually acquire target aircraft.
56.
Had the airplane been equipped with an airborne collision avoidance system that used automatic dependent surveillance broadcast in information to show directional traffic symbols.
The crew of flight 5342 would have received enhanced information about the risk posed by the helicopter which could have enabled them to take earlier action to avert the collision.
from 57.
Although the pilot and instructor pilot on board PAT 25 were equipped with tablets that had the ability to display traffic transmitting automatic dependent surveillance broadcast out.
It is unlikely that the pilots were using the tablets to monitor or identify traffic at the time of the accident due to the workload associated with lowaltitude flight.
from 58.
Technological advances since the development of traffic alert and collision avoidance system 2 operating standards may allow airborne collision avoidance system XA with reduced inhibit altitudes to have an expanded alerting envelope while reducing nuisance alerts.
Number 59.
Although not commercially available, had the helicopter been equipped with airborne collision avoidance system XR with integrated oral alerting, the the crew would have received an alert regarding flight 5342 and would have taken action to avert the collision.
Number 60.
Multiple data sources provided evidence of mid-air collision risk between fixedwing aircraft and helicopters at Ronald Reagan Washington National Airport, including on approach to runway 33 before this accident.
However, the limited access to and use of available objective and subjective proximity data hindered industry and government stakeholders ability to identify hazards and mitigate risk.
Number 61.
Improving stakeholder access to standardized and objective information about aircraft close proximity encounters for use in safety assurance processes would increase the likelihood of detecting and mitigating hazards before accidents occur.
Number 62.
the Federal Aviation Administration's lack of an established process to inform parties about their involvement in events such as near mid-air collisions or traffic alert and collision avoidance system resolution advisories reduces the likelihood of fully understanding and mitigating future midair collision risk.
Number 63.
The future.
The Federal Aviation Administration Air Traffic Organization had multiple opportunities to identify the risk of a mid-air collision between airplanes and helicopters at Ronald Reagan Washington National Airport.
However, their data analysis, safety assurance, and risk assessment processes failed to recognize and mitigate that risk.
64.
The Federal Aviation Administration Air Traffic Organization's application of its safety management system did not effectively coordinate safety assurance and safety risk management activities with external stakeholders in the Ronald Reagan Washington National Airport Class B airspace.
Number 65.
Changes to Ronald Reagan Washington National Airport air traffic control tower standard operating procedures to the accident removing the requirement for the operation supervisor OS to document the time and reason for combining or decombining the helicopter control position in the facility log made it less likely that the OS would consider and evaluate the risks associated with combining or decombining the position.
Number 66.
Safety risk management practices were not fully integrated into Ronald Reagan Washington National Airport air traffic control tower operations and did not identify or mitigate the operational challenges faced by controllers or the lack of guidance regarding operational risk assessments for controllers and supervisors.
Number 67.
Federal Aviation Administration Air Traffic Organization ATO management did not follow the tenants of safety management systems to support its workforce, encourage open communication, identify and mitigate risks or foster the adjust culture which eroded the overall safety culture within ATO.
Number 68.
The Army did not have a flight safety data monitoring program for helicopters and as a result was unaware of routine altitude exceeded and related risks in the Ronald Reagan Washington National Airport terminal area.
Number 69.
The Army's safety reporting system for pilots were not well utilized and did not provide the organization with information about close encounters between Army helicopters and other aircraft that were later found to have occurred frequently.
Number 70.
The Army's process for allocating resources to aviation safety management did not ensure the development of a robust safety management system for helicopter operations in the Washington DC area.
Number 71.
the Army lacked a positive safety culture and that its aviation safety system did not have the capacity, resources or analytical mechanisms necessary to consistently detect, interpret and act on weak signals of latent hazards.
As a result, safety assurance and organizational learning were degraded.
Additionally, this report contains the following two findings previously issued by the National Transportation Safety Board in March of 2025 in its report titled Deconlict Airplane and helicopter traffic in the vicinity of Ronald Reagan Washington National Airport.
Number 72.
Separation distances between helicopter traffic operating on Route 4 and aircraft landing on runway 33 as they ex as they existed at the time of the accident were insufficient and posed an intolerable risk to aviation safety by increasing the chances of a mid-air collision.
Number 73.
When Route 4 operations are prohibited as recommended in safety recommendation A-25-1, it is critical for public it is it is critical for public safety helicopter operations to have an alternative route for operating in and around Washington DC without increasing controller workload.
Chair, >> that was a lot.
Um, we're uh, and just for awareness for voting, there may have been a few words off here and there.
That's not a criticism, Brian.
>> We're just going to go with what's written in here, but could you pull up uh, finding 37 really quick on the screen? Yeah.
>> Okay, we're good.
We just needed to check something.
Thank you.
Okay.
Um I know we have uh a number of amendments to the findings as proposed and typically before we get to new findings and consideration of any new findings we go through any amendments to the existing findings as proposed.
So the first one I have uh from member Graham is finding 17.
>> I'm just going in order of number.
>> Yes, that's correct.
Can I I can ask the team to pull up amendment three.
That way everybody in the audience can see this.
We're online.
As you see, I have an addition in there um for finding 17 at the beginning.
So my amendment I move to add to finding 17 due to the extended time on position at the time of the collision and his expectation biased based on his prior experience at DCA ATCT and the rest is the same.
>> Is there a second? >> Second.
>> All right.
Um member Graham you're recognized.
>> Yeah.
staff presented a a compelling wellthought through case as to why the operation supervisor was likely experienced reduced alert alertness and diligence and vigilance, excuse me, due to his extended time on position.
I agree with that.
However, this investigation at every turn uncovered long-standing issues at DCA air traffic control tower.
We've heard countless times now about the make it work mentality that penetrated seemingly every facet of the workforce and how the workforce in the tower had begun to accept unsafe conditions or near misses as routine and part of the job.
Given this culture in the tower and at the time of the as accident, I believe the operation supervisor also had expectation bias given his prior experience at DCA as a controller that things would work themselves out.
In his post accident interview, for example, he recalled there was what only one helicopter on the frequency at the time when there in reality there were five.
This statement indicated that the operation supervisor was not aware of the actual helicopter traffic volume or that the local controller was being overwhelmed.
Because of this, I believe this finding must incorporate that the supervisor's expectation bias also led to his reduced alertness and loss of situational awareness.
>> Thank you.
Uh member, anything? >> I don't have any questions.
>> Okay.
Um I actually agree with everything you just said.
Uh my concern is the term expectation bias.
We did find in this uh investigation expectation bias on the part of the instructor pilot with respect to aircraft on runway lined up for runway one.
that the expectation bias was the instructor pilot had an expectation that that's where 5342 was and was staying.
The definition of expectation bias is expectation bias occurs when a pilot hears or sees something that he or she expects to hear or see rather than what actually may be occurring.
That expectation often is driven by experience or repetition.
For example, if a pilot is regularly cleared to cross a particular runway during operations at a familiar aerodome, he or she may come to expect the clearance.
This could cause a potentially dangerous situation if on a particular day the pilot actually is instructed not to cross the runway in question due to another aircraft or um uh landing or taking off.
And that's from the FAA.
So, I'm not sure that expectation bias is the right term.
But let me ask uh where um staff is on this.
>> So staff does not concur and I defer to Dr.
Wilson to explain the reason >> to be honest.
Chairwoman Hamdy, you gave a great definition of um expectation bias and we would also agree that um expectation of bias does not apply in this instance.
Can I ask you how there was something more than just being tired or fatigued? If there was like I pointed out that there was a loss of situational awareness and they could just make it work was definitely running rampant over there.
>> We definitely did see that.
I I think staff could support more um complacency instead of expectation bias that he was lulled into um thinking that everything would work out as it always has in the past.
>> So would complacency be a better word to replace in there? >> Staff would support complacency instead of expectation bias.
>> All right.
So, let uh can you pull it up on the monitor because I I agree.
Sorry.
Uh >> I I I agree with those >> friendly amendments >> statements.
>> Yeah, I'll I'll offer a a friendly amendment, but I also don't want to cut off member Enman because it looked like you wanted to add something to the discussion.
I think my only question would be and we gave staff this yesterday before 5:00 pm and if they're going to raise issues, it would have been nice to before we walked in to have this discussion to at least had those issues because we were courteous enough to try to provide them.
So that's my only frustration.
>> Well, I mean, if we're going to go there, and we can go there.
Staff has been working night and day on this report.
And member Graham, we'll get back to your amendment in a second, >> but at any time you could have provided the amendments.
What you did is give it to them at 5:00 p.m.
I said, "We are a safety agency.
We are a safety agency and I prohibit communications between board member and staff from 7:00 p.m.
to 7:00 a.m.
Why? So that they can get sleep." What you did combined was slap 35 amendments or so on them at 5:00 p.m.
Guess what? They were working till 10 or 11 p.m.
So, actually, I would like your commitment that in the future, much like the House and Senate does, that we ha we agree that all amendments will be provided to staff three business days before the board meeting.
Would you commit to that? >> I would commit to that.
Excuse me.
Are you okay with me talking? >> I provided them at 3:30.
Here's the problem.
>> I know staff continuously changed, made amendments throughout the week.
We didn't get the report, the fi our our recommendation, our staff responses until late Tuesday, less than a week.
So >> you you submitted the amendments that you already gave to staff a week ago that they already responded to objecting to.
So you resubmitted them at whatever 3:30.
You all had I don't know like what 2 hours to go through any of that.
It's not right.
3 days in advance that >> we have amendments right here.
This is part of it too.
If we have an amendment, >> that's fine and we can talk about it openly, but it's not 30 of them last minute.
>> I will let me in.
>> No, I would go back to I think if we need to revise the board order, we can certainly do that internally, but I would hang on a second.
I would please like to answer.
>> Your request to us was to submit them by five.
That was your request to us.
>> We didn't set that deadline or that time and that's the reason why we tried to and we just saw an amendment from you about an hour ago.
Uh that amendment was because we just found out that FAA changed the individual office about hours ago.
>> Okay.
>> But what I will tell you is that I said no later than five.
You knew what you were going to submit and you put staff in a very bad position.
They got no sleep last night.
We are a safety agency.
We can do better.
So do you commit to submitting amendments in the future three days beforehand? I will commit to having a discussion with the board about our procedures.
>> I will commit it.
Okay, good to know.
So, uh, let's go back to finding 17.
If you would like to withdraw your amendment, uh, member Graham, and resubmit a new amendment.
>> I would like to do that.
>> Great.
My motion is for a new amendment finding 17 17 to read due to extended time on position at the time of the collision and his complacency.
Are we okay with based on his prior experience at DCA? And the rest stays the same or just his complacency? putting me on the spot, having to think on the fly.
>> Yeah, I'm here too on the spot.
>> Um, so yes, I mean, he would get complacent based on um his experience being there, but I also think that it's um you know, we have more of the uh poor safety culture, the um prioritizing um moving traffic over safety.
So I think it might be a little bit more than just his prior experience.
Um I would be more comfortable with just putting complacency.
>> Okay.
So I I would like to amend my amendment to add after a time of collision and his complacency, the rest all remains the same.
That is my motion.
>> I second.
All right.
Uh any further comments? Great.
Uh council will uh uh call the role and tally.
>> Okay.
Board if in favor of amending finding 17 say I.
If opposed say nay.
Member Graham >> I.
>> Member Enman.
>> I.
>> Chair Homundy.
>> I.
>> All right.
Chair majority is in favor.
>> Great.
And moving to finding 20.
>> Finding 20.
Amendment four.
Please if you would pull that up.
My motion is uh I move to amend finding 20 after the positive and control instructions, to add the crew of either aircraft striking they so the crew of e of either aircraft could have taken immediate action to avert the impending collision.
That's my motion.
Second >> staff.
>> Staff concurs.
>> Agreed.
Uh council t uh tally the vote or call the role.
Tally the vote.
>> Yes chair.
This is a fa a vote in favor of amending uh finding 20.
Uh member Graham >> I.
>> Member and I.
>> Chair Hamundee.
>> I a majority in favor.
Chair.
>> Great.
Uh, amendment five, if you can pull that up.
Finding 23.
>> So, I moved to amend finding 23 to strike the word salient and add certain.
That's my amendment.
>> Uh, staff >> staff does not concur.
Staff believes that the word salient is the correct word here because it would have indicated to the crew that something was offnominal or not normal about this airplane's approach path.
>> Can I >> Yep.
>> I I don't totally agree with that because they did receive the 33 and we we can argue with that all night, but I just and I I gave my theory on that earlier.
No, I'm sticking I I I like to stay with my amendment certain with certain >> Could uh staff read what from the CVR transcript what was missed? Um I'm sorry I don't have this transcript in front of me but it was um I believe it was CRJ in the vicinity of Wilson Bridge at 1200 ft uh circling to runway 33 and it was what was missed was uh circling.
So it came across as CRJ 33 >> or 233 depending on what the partial word sounded like.
So um there we can't know for certain what would have happened if the word circling had come through in cases where um a person an operator needs to recognize that something's out of the ordinary and deserves special attention.
The more salient a cue is that draws that attention the better.
and in conversation with some of the pilots on the team uh such as Captain McKini who's a retired naval helicopter pilot.
Um it was the consensus that hearing the word that there was an aircraft on approach to one of the runways that was actually circling the airport would have drawn would have been more likely to draw the crews attention.
Um and so that's why staff uh wrote it that way.
>> Okay.
Um, I do not support the amendment.
Uh, salient meaning standing out, prominent, notable significance.
You believe circling uh could have been notable.
We don't know, but you believe that was a salient cue.
Um, I appreciate that member Graham does not.
Uh, certain though just means some, but not all.
>> Can Can I ask again? Yep.
Did you say it was a consensus amongst the whole team? >> I don't think there were any objections.
Yeah.
On the team itself.
We could certainly pull the >> We could probably argue this all night.
And I I gave you my theory before.
I I don't think this is worth the battle.
So, I'd like to withdraw my amendment here.
>> Okay.
Thank you.
>> Which means I want to withdraw uh amendment six also.
>> Okay.
Thank you so much.
would before we go on I I do have a question.
Amendment six was to 26.
Uh we will be moving uh beyond that to finding 42.
But before that comes up, I just was wondering if I could ask a question about finding 31.
Could you bring up 31? And this is just a a a question maybe for Mr.
Shahoon.
Uh and just in reading it, the term allowed by design uh instrument error.
So the tolerance stack there was in the allowable tolerance was instrument error, position error and an adjustment for temperature.
Is that all allowed by design? >> That's correct.
>> Okay, great.
Moving forward.
Just wanted to confirm that.
Moving to amendment seven.
Uh finding 42.
Member Graham, you're recognized.
So uh I move to amend 42 to read um after Washington DC comma area by Ronald Reagan Washington airport air traffic control tower the army and other helicopter operators led on from there so striking the both and the addition after army and that's my motion as you can see on the screen Second.
>> Thank you.
Staff >> staff concurs.
>> Excellent.
That was easy.
Um I also concur.
So councel.
>> Okay.
Board then.
Uh member Graham >> I.
>> Member Enman >> I.
>> Chairy >> I.
>> Majority approves.
Chair.
>> Amendment eight finding 56.
Recognize member Graham.
>> Yes.
Uh my motion is to amend finding 56 as you can see on the screen with the red lettering to add the word likely after have and before received.
>> I'll second it.
>> Thank you staff.
>> Staff does not concur.
Um and and I would point out that that as we have it written here, it says Acast that use ADSBN information to show directional traffic symbols.
So staff doesn't believe that that would likely enhance information about the risk, but it would enhance risk.
enhance information and enhance.
Exactly.
>> Uh, member Graham.
>> Um, I'm I'm Is this is this consensus of the staff? >> Yes, it is.
>> Okay.
I I I'm willing to remove uh withdraw my amendment at this point.
>> It's not worth it.
>> Thank you.
Uh, amendment nine, finding 59.
Um, my motion is I move to amend finding 59 to and in this case I I am going to stick with removing the word would in both locations to could and that's my amendment.
>> Is there a second? >> Second.
>> Would the uh gentleman consider a friendly amendment? I know we talked earlier, but at this point I am this technology is not out there.
It's in simulation.
I know they're working on performance standards, but it's not there.
And I know there's a difference in the world from simulation and development to actual I I work for an aircraft manufacturer actually two of them for a lot of my career and things don't always work out the way you want them to or you think they theoretically will.
So, I would I'm going to stick with code on both.
>> Thank you.
Uh staff.
>> So, we're about halfway there.
Um staff believes that uh the first one the wood because we did show that in simulation that that should remain.
However, we agree with you on the second one to make it a could.
>> I understand that, but I'm still sticking with could on the front one for my explanation.
Can uh can I ask staff to talk about the simulation and what you saw in the simulation? >> Absolutely.
We'll defer to uh Mr.
Wagner.
As we demonstrated in our presentations, simulation using the accident flight data um with the developmental version of AKS XR did um show improved alerting.
M >> for the helicopter.
>> I I understand that and like I said that this is still simulation.
We're still somewhat in a theoretical world and uh this has been a long time coming and it's not there yet.
So I still think it's a big could, not a would.
Thank you for that though.
>> All right.
Um I disagree.
I believe we have uh recommended numerous technologies over the years that were still in development and we were very clear as recently as March 2024 uh to approve uh uh technology for under 20 miles per hour on positive train control which is not designed right now for under 20 20 m per hour.
and we concluded that it would have prevented an accident but uh understood.
So we'll uh uh council you'll call the role and tally the vote.
>> All right chair uh member Graham >> I.
>> Member Enman >> I.
>> And chair HD >> no.
>> All right chair majority is in favor.
>> Thank you.
Uh we're going to pull up amendment 10 finding 63.
So my uh motion is to amend finding 623 to read.
The FAA atto was made aware of and had multiple opportunities and the rest remain the same.
So the ad was made aware of comma and >> there a second >> second >> staff >> staff concurs.
Well, I mean, do you want to talk about it? Pretty obvious.
>> Yeah.
I also support the amendment.
Uh Tom.
>> All right.
Member Graham.
>> I.
>> Member Enman.
>> Hi.
>> Chair Hamundy.
>> I.
>> Majority is in favor.
Chair.
>> Amendment 11.
Finding 71.
>> Yeah.
I just uh uh my motion is to amend finding 771 to read the Army's aviation safety system failed to consistently detect, interpret, and act on signals of latent hazards resulting in degraded safety assurance, organizational learning, and safety culture.
That's my motion.
Second discussion.
>> Second.
>> We change it again.
>> Second.
>> Second.
>> Oh, I already seconded.
We're good.
It's time for discussion.
Uh, member Graham.
>> Um, I it's just cleaning it up a little bit.
It's It's pretty much I I thought it was very wordy to begin with.
>> Okay.
Um, >> just I would refer to staff if you want to go further.
I can talk more about it.
>> Uh, we have no technical objection.
>> I also support the amendment.
So, we can just go to vote, I think.
>> All right.
Member Graham >> I.
>> Member Enman.
>> I, Chair Hammond, >> I.
>> All right.
Majority in favor.
>> Excellent.
Moving on to amendments uh of new findings.
>> I have two.
>> Yes.
I'm sorry for everybody here.
>> Graham is >> talking just as much as Mr.
Curtis here.
>> Go ahead.
>> It's not over yet.
Uh amendment one, please, if you'd pull that up on the screen.
So, this is a new amendment uh that I I move to add.
It reads, "In the two minutes before the accident, when traffic volume was increasing, the assistant local controller should have prioritized surveillance of aircraft in the air in order to assist the local controller rather than diverting her attention to the lower priority task of documenting helicopter information, which could have been completed when traffic volume and complexity had subsided." That's my motion.
>> Second.
Uh time for discussion.
Member Graham.
>> Yeah, as I referenced in my opening statement, there were individual shortcomings comingings that contributed to this accident that need to be pointed out.
As stated in the report and earlier in our discussion, the primary duty of the assistant local controller is to maintain surveillance of the local traffic pattern and landing area and alert the local controller of any unusual situations or traffic conflict.
These duties and responsibilities are clearly stated in the DCA air traffic control standard operating procedures.
The fact is the the ALC was completing other tasks in the leadup to and at the time of the accident that should have been a far lower priority.
None of the current findings as drafted acknowledge this situation and I believe this report would be incomplete without a reference to the assistant local controller.
We already have multiple findings related to the local controller and the operations supervisor and this propose proposed finding as drafted was included in staff's initial report just not as a finding.
I also want to note that finding 16 is currently drafted does not or I'm sorry does acknowledge that had the HC and LC positions been staffed separately it would have allowed PAT25 to receive a more timely and effective traffic advisory because both the local and the assistant local controllers could have had increased ability to recognize the pending collision.
However, while I believe that is true, that finding does not in any way reflect that the assistant local controller was not appropriately priorit prioritizing her duties on the night of the accident.
>> Member Inman, anything? >> I have no comments.
>> Okay.
Um so I support the amendment.
Uh as we lay out uh in our uh draft report, the assistant local control controller uh their duties are number one to alert the local control controller of any unusual situation or traffic conflicts.
Number two, maintain surveillance of the local traffic pattern and landing area.
Number three, assist the local controller control controller with monitoring of aircraft on final.
So I support uh the amendment staff.
So staff does not concur.
staff would prefer to retain the system focused finding proposed in finding 16 rather than adding this finding which focuses on an individual's actions.
And I'm just going to read 16 really quick for public transparency.
That is the one I believe you just read, which was had the helicopter and local control positions been staffed separately, PAT 25 might have received a more timely and effective traffic advisory, >> but it wasn't.
So, it's a binding.
>> Yep.
All right.
Any other questions? Uh Tom, you're up again.
>> All right.
This is adding a finding by member Graham.
Uh, member Graham >> I.
>> Member Enman >> I.
>> Chair Hammond >> I.
>> Uh, all in favor? >> Member Graham up again.
>> Amendment two.
It's my last one for right now.
I appreciate everybody's patience.
Uh, that's that are here.
I I move to add new finding.
The local controller and helicopter controller positions should have been separated at the time of the accident given traffic volume and complexity.
>> Second.
>> A second.
>> Uh discussion.
>> Yeah.
If uh as I previous mentioned, I agree with finding 16 is currently drafted, which states that had the helicopter and local control positions been staffed separately, PAT25 might have received a more timely and effect effective traffic advisory.
In my view, that finding is missing a company finding making it clear that the two positions should have been separated given the traffic volume and complexity of DCA in the leadup to the accident.
based on the standard operating procedures and advide and available guidance from the operation supervisor to decide whether or not to combine or decombine the position.
It is abundantly clear to me that the complex mix and volume of traffic that night warranted separating the positions.
Additionally, we have heard earlier during our discussion the tower had enough staff available to staff the positions separately.
Given these realities and staff supported this statement in our in our earlier discussion, I believe the board is in a place with this evidence presented to make the determination that the positions should have been decombined.
>> I support the amendment staff.
>> So staff doesn't necessarily disagree uh with that.
Um, but we do think that it's found in in finding 15, 16, and 22.
And I do have those if it would be helpful for me to read them.
I I honestly think we tend to talk around it a lot, but we don't specifically state it.
And I think it's worth a positive statement as a finding.
That's my position.
>> Okay.
Um I I agree.
Uh I think 15 really focuses on controller workload as a result of the combining of the helicopter control and local control positions.
16 talks about um you know if it had been staffed separately, PAT 25 might have received a more timely alert and I know you mentioned another one, but I do think we have to be very clear that they just should have been separated.
>> I tend to agree the the evidence is very evident.
Even the controller said he felt overwhelmed at one point and then it got better, but then 90 seconds before it was higher than when he thought he was being overwhelmed.
>> Yep.
>> There's no doubt there's evidence that points to it.
>> All right.
Tom.
>> All right.
Board on um adding a new finding.
Member Graham.
>> I.
>> Member Enman.
>> Hi.
>> Chair Hammond.
>> I.
>> All right.
Uh all in favor.
>> All right.
Member Graham, you are done with that.
All right.
Moving to uh member Inman.
>> Thank you.
>> Your proposed proposed finding.
It's a it's just a new finding, not amending finding one.
>> Correct.
>> Okay.
Can you pull up? Oh, do we have that? >> Yeah.
Okay.
So my proposal is to actually new finding.
Uh the Metropolitan Washington Airport Authority crash fire rescue along with all other responding units from various jurisdiction within proximity to the airport demonstrated exceptional readiness and coordination arriving on the water within approximately 4 minutes of the accident.
This response time reflects a model operational capability and contributed positively to incident management efforts.
Second >> member Inman, you're recognized.
>> I I point back to it's so often that we have to talk about the faults of people and things that are being done and and I think back, this is actually the 44th anniversary was a few weeks ago, an Air Florida crash that occurred on January 13th in 1982 when 70 people perished.
And unfortunately, we're back here again 44 years later.
But during that investigation, we found that the Metropolitan Washington Airport Authority and other responders did not have a cohesive system.
They did not coordinate.
In fact, the ARF firefighters were not even notified until 3 minutes after the crash occurred.
And so, whenever there is a positive contribution, we should at least acknowledge that uh at least the airport authority and other responders had worked hard in the event that this happened.
And while unfortunately there were no survivors, the the likelihood of a recovery was much greater based on their immediate action.
They rushed into the ice to try to save people.
Was there a second? Okay.
Member member Graham, do you have a comment? >> I totally agree with the member Inman and I I'm supportive of this.
Uh I am supportive of the intent but I am not supportive of the amendment.
Uh there were a number of response agencies when the NTSB and I will refer to Wamada in the Lant Plaza smoke accident east Palestine Dupant.
uh we uh look at we are aware very early that there are major issues which is why we form an emergency response group to look at the emergency response and take lessons learned that's our mandate lessons learned and turn those into lives saved.
Uh so we did delve deeply on emergency response and many accident investigations including those I referenced.
Uh highways a recent one we investigated looked at emergency response because they were had some questions early on but it turned out that it was timely and adequate.
And so the res what we put in uh reports over and over again is emergency response was timely and adequate.
Uh, I will say for this one, uh, not only did we have Metropolitan Washington Airports Authority, which did a wonderful job, but it was the DC fire chief who was the incident commander because they have jurisdiction over the river uh to the Virginia shoreline.
And certainly uh they brought in the DC fire chief brought in the National Capital Region Incident Magnet Team to manage and coordinate unified command.
And while there was a uh four minutes from the crash phone that rang from the tower to a boat on the water, which is admirable, I have concerns about issuing a finding when we did not.
We have zero evidence of looking at that in our docket.
We are an evidence-based organization.
When I look at survival factors, uh, which would have looked at the emergency response and I looked at the survival factors report.
Uh it goes into a great deal of information including cabin interior wreckage, loose portable wreckage transported by divers, small boat recovery, fuselage and wreckage obtained during crane and barge recovery.
We looked at seat seats restraints, other survival equipment.
We looked at emergency equipment and placards on board the aircraft.
We looked at helicopter configuration and seating.
We looked at the helmets to see is there anything any lessons that we could have learned there.
We did not.
However, we did document an overview of the report and we gave an emergency response summary and noted applicable regulations to how to conduct emergency response, but we did not evaluate the emergency response.
And so the only analysis that was actually conducted by that group was really focused on survivability and whether it was survivable which it was not.
There is no analysis.
And so when we uh adopt a finding, we need to have evidence to support that finding.
There wasn't an analysis.
There wasn't a factf finding portion of the investigation.
uh re with respect uh particularly particular to incident command unified command or emergency response in this case it's not drafted in a way that is meets the high standards of our organization crashfire slrescue and then we go into demonstrated exceptional readiness and coordination.
We don't know that.
We did not evaluate that.
And while their work is heroic, truly heroic, again, we are evidence and fact-based.
And we do not have that evidence in our docket.
And so when we say things like this response time reflects a model operational capability, we have not evaluated whether it's a model.
We have no factual evidence to look at any other model across the United States and compare it to.
So I do not support the amendment staff.
>> Thank you, Chair.
I I've waffled on this and and I I fully appreciate the intent and I think every >> yes >> uh every one of us um that were down uh at the airport that night and that week appreciates everything that the first responders had done.
And so I had asked staff go back look to see what we can support.
And in in all reality, we had put something together um to try to come back and say, "Hey, here's something we could support." But when I looked at the chair's uh opening presentation and I was here counting, there's 29 different organizations that were present and we we can't support that.
You know, we didn't look at the response for all of them.
So, um, so I don't concur, um, or staff doesn't concur.
And, and again, it's not that we don't think they did a fantastic job.
We've got to go back to what we did indeed investigate for our findings.
And in this case, we didn't look at the first responders as a whole.
>> There was a I think it was handed to me.
Did staff have an alternative version? Yeah, and that's what I was saying.
I' I've got one that I've worked on, but I'm not even going to propose it now because it it it talks about one organization.
There were 29 different organizations.
>> I would I I I respectfully I I do agree with you.
Um perhaps we could if if member Enman is is willing to cons to withdraw that and consider the staff finding withdraw his amendment and consider the staff finding perhaps we could just say and other response organizations in there.
>> That's the reason I asked if the staff had an alternative.
>> So um >> I think we got a slide if that would be helpful.
>> How about a slide? Yep.
And again, I I Oh, sorry.
I'm just I don't think that's >> okay.
>> While he's doing it, would you like me to read it? >> Actually, I'll tell you what.
I'll make it easy for you because somebody printed this earlier.
That's the reason I was asking.
Um, I'm just going to try to read this so hopefully catch up with it.
The Metropolitan Washington Airport Ports Authority, airport rescue and firefighting and airport operations staff responded immediately and in accordance with applicable emergency plans and regulatory requirements, rapidly establishing incident command, deploying land and water-based resources, and coordinating mutual aid under complex nighttime and onwater conditions.
>> Does that sound >> That's what I have.
Yes, >> I have no problem amending to that.
Obviously, the intent was to try to make sure we at least say the appreciation for the other first responders and I hope it at least can be noted in some way.
And I apologize, this was a miss on my behalf during our staff meetings.
I was we didn't have enough time to get to that to where we was trying to get it into the investigative report, but obviously the docket needs to support it as well.
>> I'll offer that as an an amendment to my own amendment or a new amendment.
second.
Can you give me two seconds or what? You going to put it up? Yeah.
I think the only part I have a concern with is the words rapidly establishing incident command because it was the D DC fire and rescue who did that.
So would you consider striking that and leaving the rest because DC fire didn't do the other things.
So if we add them, >> I I would tend to disagree.
And the reason I say I tend to disagree is that had a an incident command trailer on runway 33 that night and they established their own incident command.
That may not be the unified incident command.
>> It's not unified command.
I check I checked with >> right our direct >> Yeah, >> I checked with our director of special operations this morning on who established incident command immediately afterwards.
They had an incident command trailer.
They did not establish incident command.
DC Fire and Rescue was also immediately there and established incident command and it was transferred to unified command the next morning.
>> Okay.
>> Okay.
>> What I'm saying is that would be inaccurate to say that Mwah rapidly >> if you'd like to strike it, we can.
>> Oh, okay.
Got it.
>> You said okay.
>> Oh, I I I >> I don't know if you're getting used to that.
We will say okay some.
No, no.
I didn't know.
I didn't know if you were just saying I heard you.
Yeah.
Okay.
So, let's pull this back up.
>> So, I'll repropose it for the third time.
The Metropolitan Washington Airports Authority, airport rescue and firefighting and airports operation staff responded immediately and in accordance with applicable emergency plans and regulatory requirements, deploying land and water-based resources and coordinating mutual aid under complex nighttime and onwater conditions.
Second.
I'm seconding it.
>> Okay.
>> Okay.
Uh, anything else? Okay, Tom.
>> All right.
Uh, a vote in favor of the new finding.
Member Graham.
>> I.
>> Member Enman.
>> I.
>> Chair Holiday.
>> I.
>> All in favor? >> Excellent.
Member Mman, you're recognized.
>> Amendment uh proposed additional finding.
Number two, the accident could have been prevented if the instructor pilot had ended the NVG proficiency flight at numerous points due to pilot performance and meteorological conditions while on NVGs.
>> Second.
Um, okay.
Staff, >> thank you.
Staff does not concur.
Um, we could have said this about a a number of things.
I mean, we could have said if if the helicopter would have never taken off, the accident would not have happened.
Um, in addition, the instructor pilot did not consider the pilot's performance as poor enough to discontinue the flight.
And in reality, a check uh ride is a training opportunity.
And even if it's unsuccessful, um that's not a reason to discontinue the flight.
Are you Oh, well, you were getting a lifesaver.
I thought you were trying to reach for your microphone.
Okay.
So, uh I agree with staff staff's assessment.
I uh do not support the finding.
The finding says the accident could have been prevented if the instructor pilot had ended the NVG proficiency flight at numerous points due to pilot performance.
That's making a uh judgment and assumption on pilot performance that is nowhere discussed in the report.
Uh Mr.
McKenna, I would ask uh do you believe is there any way to know whether the pilot uh w would have passed uh the standardization test? >> We do not know what the instructor pilot would have determined at the end of the flight because they hadn't completed the flight yet.
Did you determine that there was any deficiency with the pilot performance? >> We found no deficiencies with the pilot performance with the information that we had.
>> And was there any reason why uh the instructor pilot should have ended the NVG proficiency flight due to meteorological conditions? It was windy but gusty.
>> Yeah.
Uh the winds I don't think affected the MVG portion of the flight.
Um I I have been thinking about this a little bit more and I would say that um the helicopter the crew was transiting route one and four which was basically they were repositioning down to Davidson airfield to do more of the of the proficiency check.
So this was just a transit portion of the flight um which any helicopter crew would have been doing at any time along a published route and even if they had decided to return early they probably would have transited that route anyway.
So it still would have put the helicopter in jeopardy of the intersection of uh route 4 and the approach path to runway 33.
Did we anywhere in the report uh Mr.
McKenna uh or Mr.
Banning uh talk about pilot performance at numerous points of the flight? >> I don't recall that we we have anything on that.
>> We did not.
So, this would be a finding with no backup information in the report and frankly would put staff in the position of making it up, I guess, because that was not your assessment.
Would member Inman consider withdrawing this? I I I actually will, but on a different argument, I guess, because the CVR itself reveals, we talked about this earlier, a missed turn, a failed landing, um 27 different instances of of wind buffeting and things like that, but actually hearing your latest reasoning, had they even called it off, they still would have had to transition back to Fort Bel.
That to me sounds more rational.
Um, and I can understand that a little bit more.
Even if you said no, you still got to get home.
you got to land the aircraft and you can't take a different route than what was already there.
So with that, I'm willing to withdraw uh on that premise.
>> Okay.
Are there any other amendments to the findings? >> All right.
Moving on to the probable cause.
Mr.
>> Oh, wait.
We actually have to approve the propo the findings as amended.
>> I will move for that.
>> I will second.
Any further discussion? All right.
This is on findings as amended.
Uh Mr.
McMurray.
>> That's it.
Chair.
Uh member Graham.
>> I.
>> Member Enman.
Hi.
>> Chair Hamdy.
>> Hi.
>> All in favor? >> Mr.
Curtis.
>> Mr.
Curtis >> chair staff proposes the following >> I am reminded that staff would like a brief break so sorry uh so we will do 15 minutes five uh 540 okay or you want 5:45 20 minutes.
Okay, we are back.
Mr.
Curtis, can you read the proposed probable cause? >> Certainly.
Chair staff proposes the following probable cause.
The NTSB determines that the probable cause of this accident was the placement of a helicopter route in close proximity to a runway approach path with no procedural mitigations to separate helicopter and fixed wing air traffic fixed wing traffic.
the FAA's inadequate evaluation and review of data that indicated the risk of a mid-air collision in that area and their failure to act on recommendations to mitigate that risk.
Also causal was the air traffic air traffic systems over reliance on visual separation in order to promote efficient traffic flow without consideration for the limitations of the see and avoid concept.
and the Army's lack of a fully implemented safety management system which should have identified and addressed hazards associated with altitude exceeded on the Washington DC helicopter routes.
These systemic failures allowed for a mid-air collision after the helicopter crew was unable to apply visual separation due to the misidentification of the location of the airplane.
Contributing factors include the limitations of the traffic awareness and collision alerting systems on both aircraft which precluded effective alerting of the impending collision to the flight crews.
the high workload of the controllers due to the combination of the local and helicopter control positions and the absence of effective data sharing and analysis among the FAA aircraft operators and other relevant organizations.
Chair, >> thank you.
Well, we all had some changes, so I'm going to recognize member Graham.
Thank you, chair, and I appreciate staff and the individual members are working individually with me on this.
Uh uh I I move to amend the probable cause to read and I'm not even going to write well I guess you got a red line in there.
That's pretty good.
I'm just going to read the whole thing.
Uh because I can't read through those red lines right now.
the NTSB.
My uh my motion is uh to amend the uh the probable cause to read, the NTSB determines that the probable cause of this accident was the FAA's placement of a helicopter route in close proximity to a runway approach path, their failure failure to regularly review and evaluate helicopter routes and available data, and their failure to act on recommendations to mitigate the risk of a mid-air collision near Ronald Reagan, Washington ational airport as well as the air traffic systems over reliance on visual separation in order to promote efficient traffic flow without consideration for the limitations of the see and avoid concept.
Also causal was the lack of effective pilot applied visual separation by the helicopter crew which resulted in a mid-air collision.
Additional causal factors were were the tower team's loss of situational awareness and degraded performance due to a high workload of the combined helicopter and local control positions and the absence of a risk assessment process to identify and mitigate real-time operational risk factors which resulted in mis prioritization of duties.
inadequate traffic advisory advisories and the lack of safety alerts to both flight crews.
Also causal was the Army's failure to ensure pilots were aware of the effects of air tolerances on barometric altimeters in their helicopters which resulted in the crew flying above the maximum published helicopter route altitude.
Contributing factors include the limitations of the traffic awareness and collision alerting systems on both aircraft which precluded effective alerting of the impending collision to the flight crews.
an unsustainable airport arrival rate, increasing traffic volume with a changing fleet mix, and airline scheduling practices at DCA, which regularly strain the DCA ATCT workforce and degraded safety over time.
the Army's lack of a fully implemented safety management system, which should have identified and addressed hazards associated with altitude exceedences on the Washington DC helicopter routes.
the FAA's failure across multiple organizations to implement previous NTSB recommendations, including ADSBN, and to follow and fully integrate its established safety management system, which should have led to several organizational and operational changes based on previously identified risk that were known to management and the absent of effective data sharing and analysis among the FAA, aircraft, OP operators and other relevant organizations.
>> Second.
I'm seconding that.
I'm sure >> I would just clarify.
I think from what was shown on the screen was not what member Graham was reading.
>> So, I just want to make sure people realize that there is some modifications especially in the front of the first that >> Yeah.
There >> is it possible to get that on the screen? >> Did they get that email? They're working on it.
>> Yeah.
Well, we'll chat about it while we if you guys can get even if you want to put it up on Word right now.
>> It doesn't have to be pretty.
>> It doesn't have to be pretty.
It was different.
That would be fine.
>> And I appreciate everybody working with me on this and I appreciate the patience of everybody here on this.
Um >> I think it's important that we we get this right.
>> Absolutely.
>> And um appreciate working with each member up here and the staff.
I know we've been round and about on this for more than one day >> and I would offer if this is adopted I can withdraw my changes to the probable cause that I had I think staff working with member Graham said can we put them all into one package so as long as we don't have an issue I won't have anything else beyond addition so could we find out if staff agrees with this can we get it okay >> if you say no >> staff concurs >> staff still employed >> I appreciate I I appreciate that.
Uh it did uh address a number of of uh my concerns.
Number one, this helicopter route shouldn't have been there in the first place.
This was terrible uh design of the airspace.
And from my perspective, I also uh felt that it was critical uh like member Inman to talk about the error tolerances on barometric altimeters and the lack of pilot knowledge of knowing that they thought they were at 200 ft and they were actually at uh 300 or above at some points uh which is unacceptable.
Um and you know I do appreciate thank you for the addition of the failure of uh multiple organizations to implement previous NTSB recommendations because we have issued a number of recommendations for years uh including to address C and Avoid and ADSBN.
So thank you for also including that.
Um, I think this is a a really good probable cause and you're right, we needed to get it right.
How about we move to a vote? If we need to hand out copies, we could just photocopy, right? Okay.
>> You ready, chair? >> You ready? Yeah.
All right.
Go for it.
>> Okay.
This is a vote in favor of the amendment to the probable cause read by member Graham.
Member Graham.
>> I.
>> Member Enman.
>> I.
>> Chair Homundy.
>> I.
>> All right.
Chair.
Majority's in favor.
>> Great.
And I don't believe we have to do anything else on the probable cause because it was an entire replacement.
Excellent.
We may now move to the recommendations.
Mr.
Curtis, you can read the recommendations.
As a result of this investigation, staff proposes the following 48 new safety recommendations, 32 to the Federal Aviation Administration.
Number one, develop and implement time on position limitations for supervisory air traffic control personnel, including guidance for district and facility level management to adapt these limitations to account for their own staffing and local standard operating procedures.
Number two, develop instructor-led scenario-based training on threat and error management that trains controllers to continuously monitor their environment to more quickly and accurately identify threats.
Promote team communication to ensure that communications are clear, timely, and assertive.
Emphasize effective scanning habits.
Recognize patterns in the development of adverse events and enhance decision-making under stress by developing habits that balance procedural compliance with problem solving to mitigate the risks of threats and errors and provide this training to all air traffic controllers annually.
Number three, develop and implement a risk assessment tool for supervisors that incorporates the principles of threat and error management to assist in risk identification, mitigation, and operational decisionmaking.
Number four, initiate rulemaking in 14 code of federal regulations part 93 subp part K highdensity traffic airports that prescribes air carrier operation limitations at DCA in 30 minute periods similar to those imposed at LaGuardia airport to ensure that the airport does not exceed capacity and to mitigate inconsistent air carrier scheduling practices.
Number five, fully implement operational use of the timebased flow management system at PTOAC consolidated terminal radar approach control and its associated air traffic control towers.
Number six, reassess the Ronald Reagan Washington National Airport's airport arrival rate with special consideration to its airspace complexity, airfield limitations, mixed fleet operations, and traffic volume.
Number seven, define objective criteria for the deter for the determination of air traffic facility levels considering traffic and airspace volume, operational factors unique to each facility, and cost of living.
Number eight, using the CR criteria established by safety recommendation seven, determine whether the classification of the Ronald Reagan Washington National Airport's a air traffic control tower as a level 9 facility appropriately reflects the complexity of its operations.
Number nine, conduct a comprehensive evaluation in conjunction with local operators to determine the overall safety benefits and risks to requiring all aircraft to use the same frequency when the helicopter and local positions are combined in the Ronald Reagan Washington National Airport air traffic control tower.
Number 10, implement anti-blocking technology that will alert controllers and/or flight crews to potentially blocked transmissions when simultaneously broadcasting occurs.
Number 11, develop and implement improvements to the conflict alert system to provide more salient and meaningful alerts to controllers based on the severity of the conflict triggering the alert.
or 12.
Once the improvements to the conflict alert system discussed in safety recommendation 11 are implemented, provide training to controllers on its use.
Number 13, revise the air traffic organization's initial event response procedures so that an appropriate on-site supervisor makes each post accident and post incident drug and alcohol testing determination based on their assessment of whether the event meets testing criteria and which controllers had duties pertaining to the involved aircraft without needing to wait for investigation or approval.
Number 14, at least annually provide training on the revised post accident and post incident drug and alcohol testing determination procedure discussed in safety recommendation 13 to all staff who have responsibilities under that procedure.
This training should include a post-arning knowledge assessment.
Number 15, ensure that annual reviews of helicopter route charts are being conducted throughout the National Airspace System as required by Federal Aviation Administration order.
Number 16, conduct a safety risk management process to evaluate whether modifications to the remaining helicopter route structure in the vicinity of Ro Ronald Reagan Washington National Airport are necessary to safely deconlict helicopter and fixed wing traffic and provide the results to the National Transportation Safety Board.
Number 17, amend your helicopter route design criteria and approval process to ensure that current and future route designs or design changes provide vertical separation from airport approach and departure paths.
Number 18.
Based on the criteria and approval process established by safety recommendation 17, review all existing helicopter routes to ensure alignment with these updated criteria.
Number 19.
Incorporate the lateral location and published altitudes of helicopter routes onto all instrument and visual approach and departure procedures to provide necessary situation awareness to fixed wing operators of the risk of helicopter traffic operating in their vicinity.
Number 20.
Modify airborne collision avoidance system traffic advisory oral alerts to include clock position, relative altitude, range, and vertical tendency.
Number 21 require existing and new traffic alerting and collision avoidance system TCAST one, TCAST 2, and airborne collision avoidance system X installations to integrate directional traffic symbols.
Number 22 require all aircraft operating in airspace where automatic dependent surveillance broadcast ADSB out is required to also be equipped with ADSB in with a cockpit display of traffic information that is configured to provide alerting audible to the pilot and/or flight crew.
Number 23 require the use of the appropriate variant of airborne collision avoidance system X on new production aircraft that are subject to traffic alert and collision avoidance system equipage regulations.
Number 24, require existing aircraft that are subject to traffic alert and collision avoidance system equipage regulations be retrofitted with the appropriate variant of airborne collision avoidance system X.
Number 25.
Evaluate the feasibility of decreasing the traffic advisory and resolution advisory inhibit altitudes in airborne collision avoidance system XA to enable improved alerting throughout more of the flight envelope.
Number 26.
If the evaluation resulting from safety recommendation 25 finds that the inhibit altitudes can be safely decreased require retrofitting of the applicable airborne collision avoidance system X variant incorporating the reduced traffic advisory and resolution advisory inhibit altitudes on all aircraft that are subject to traffic alert and collision avoidance system and equipage regulations.
Number 27 require that all rotor craft operating in class B airspace be equipped with airborne collision avoidance system AAS XR technology once the AAS XR standard has been published.
Number 28.
Create an objective definition of close proximity encounter and a public database of those encounters and their locations that can be used to monitor their prevalence and identify areas of potential traffic conflict for safety assurance and safety risk management.
Number 29.
Develop and implement a process that will in a timely manner notify involved parties after events such as near midair collisions or traffic alert and collision avoidance system resolution advisory activations such that notification occurs while relevant data remain available and before meaningful safety analysis, reporting or corrective action is no longer practical.
Number 30.
Direct an audit of the air traffic organization ATO by a competent authority outside of the ATO of AT atto safety management system functions and data sharing activities at all air traffic control facilities and determine whether these activities are conducted in collaboration with all relevant external stakeholders ensuring that the audits results are documented, reported to the administrator and made available to the public.
Number 31.
Based on the results of the audit completed in accordance with safety recommendation 30, ensure that all safety management system functions and data sharing activities at all air traffic control facilities are conducted in collaboration with all relevant external stakeholders.
Number 32.
Establish a requirement across all air traffic control tower standard operating procedures that the operations supervisor OS or controller in charge CIC document in the daily facility log that when any control position is combined with the local control position or when the OSCIC position is combined with a control position along with a rationale for doing so.
Eight recommendations to the US Army.
Number 33, revise training procedures for flight crews assign assigned to operate in the Washington DC area to ensure that they receive initial and recurrent training on fixedwing operations at Ronald Reagan Washington National Airport, including approach and departure pass, runway configurations, and the interaction of those traffic flows with published helicopter routes.
Number 34.
Develop and implement a recurring procedure at an interval not to exceed 18 months to verify the continued accuracy of recorded flight data.
Number 35.
Incorporate information within the appropriate o operator's manual for all applicable aircraft on the potential total error allowed by design that could occur in flight on an otherwise airworthy barometric altimeter including the increased position error associated with the external stores support system configuration.
Number 36.
develop and implement a transponder inspection pro procedure on all aircraft with transponders capable of transmitting mode S and automatic dependent surveillance broadcast ADSB and operated in the national airspace system NAS at least annually and upon each aircraft's entry into service in the NAS that ensures one the transponder ADSB settings are correct two the transponder is transmitting ADSB and three the transponder is transmitting the correctly assigned address.
Number 37, establish a flight data monitoring program for rotary wing aircraft the US Army operates in the national airspace system.
Number 38.
Survey US Army helicopter pilots to identify barriers to the to the utilization of flight safety reporting systems.
Develop a plan to address the identified barriers and implement that plan across Army aviation units.
Number 39.
Revise the method for allocating resources to ensure the development of a robust safety management system that will at a minimum identify and monitor the potential for mid-air collisions between Army aircraft and civil air traffic operating in the national airspace system.
Number 40.
develop and maintain a flight safety management capability that is independently resourced and functionally separate from its occupational and environmental health management system and ensure that this capability is both culturally and functionally integrated with units conducting sustained flight operations in the national airspace system.
Five recommendations to the Department of War Policy Board on Federal Aviation.
Number 41, conduct a study to evaluate the quality of radio transmissions and reception for those aircraft operated within the national airspace system to identify factors that degrade communications, equipment performance, and adversely affect the safety of civilian and military flight operations.
or 42.
Implement appropriate enhancements based on the findings of the study recommended in safety recommendation 41 to remediate identified deficiencies in air ground radio communications performance.
Number 43 require the Department of War to verify on all aircraft with transponders capable of transmitting mode S and automatic dependent surveillance broadcast ADSB and operated in the National Airspace System NAS at least annually and upon each aircraft's entry into service in the NAS that one the transponder ADSB settings are correct two the transponder is transmitting ADSB B and three, the transponder is transmitting the correctly assigned address.
Number 44, require armed services to amend their operational procedures to allow flight crews to enable automatic dependent surveillance broadcast out while in flight.
Number 45.
require all military aircraft operating in the National Airspace System, NAS, to be equipped with automatic dependent surveillance broadcast, ADSB, in with a cockpit display of traffic information that is configured to provide alerting audible to the pilot and/or flight crew and that such requirement apply whenever in the NAS the Federal Aviation Administration requires any aircraft to operate with ADSB Vote two recommendations to the Department of Transportation.
Number 46 require the Federal Aviation Administration to demonstrate at least annually that each air traffic control facility it operates has the routine capability to accomplish required postac and post incident drug and alcohol testing within the US Department of Transportation's specified time frames of two hours for alcohol and four hours for drugs and implement a process to ensure that any facility without such capability will demonstrate timely remediation.
Number 47.
Work with the Federal Aviation Administration FAA administrator to convene an independent panel to conduct a comprehensive review of the safety culture within the FAA's air traffic organization ATO and use the findings to enhance the ATO's existing safety management system and integrate it into all levels of the organization.
and one recommendation to the RTCA program management committee.
Number 48, finalize and publish the minimum operational performance standards for airborne collision avoidance system XR for rotocraft.
Additionally, this report contains the following two safety recommendations previously issued by the National Transportation Safety Board in March of 2025 in its report titled deconlict airplane and helicopter traffic in the vicinity of Ronald Reagan Washington National Airport to the Federal Aviation Administration.
Urgent recommendation A-25-1 prohibit operations on helicopter route 4 between Hannes Point and the Wilson Bridge when runways 15 and 33 are being used for departures and arrivals respectively at Ronald Reagan Washington National Airport DCA.
and urgent recommendation A-25-2 designate an alternative helicopter route that can be used to facilitate travel between Hannes Point and the Wilson Bridge when that segment of Route 4 is closed.
Chair, >> thank you, Mr.
Curtis.
Uh, okay.
We have some amendments and we'll do the same thing in order of um uh number of uh the amendment as it's in the draft report.
I have one amendment.
I think mine comes up first for recommendation 30.
My recommendation 30.
Yeah, I'm first.
Yeah.
All right.
So, uh, right now recommendation 30, uh, states that the FAA, uh, should direct an audit of, um, ATO by a competent authority outside the ATO of the ATO's safety management system.
Uh, this would be an amendment.
I offer an amendment to change that to a recommendation to the US Department of Transportation, Office of Inspector General.
Is there a second? >> Yes.
>> Thank you.
Um, first staff position.
>> Well, I'm just a little curious as to why.
>> Yes, I will tell you why.
Okay.
Um All right.
So, uh, the reason I I was okay with it being an independent, uh, audit, uh, directed to the FAA to allow an entity outside of the ATO, but within the FAA to look at the ATO and their safety management system until I we all learned a few hours ago that the Mr.
Fuller left the ATO and is now the head of the SMS office at FAA.
I'm think that creates a conflict and so uh >> staff concurs.
>> I don't even have to finish.
Okay, there you go.
>> Perfect.
Any other comments? >> Vote.
>> Vote.
>> All right, board.
Uh this is in favor of amendment to recommendation 30.
Member Graham >> I.
Member Enman.
>> Yes.
>> Chairman Hamdy.
>> I.
>> All in favor? Chair.
>> Great.
Um, next is I'm finding it.
I'm going to, uh, turn to member Graham.
>> Thank you, uh, chair.
Uh, I have three new recommendations to the FAA.
So, I'll go one at a time.
If you could call up, uh, amendment 14, please.
So, I'll go ahead and start right here.
We go.
Uh, I move to uh for a new recommendation of the FAA revise FAA order job order 7110 decimal 65 section 2-1-6 safety alert to require controllers assert positive control by issuing an alert with appropriate traffic advisory information and offering pilots an alternate course of action followed by the word immediately in quotes if they observe an aircraft is in unsafe proximity to another aircraft.
That's my motion.
>> Is there a second? >> Second.
>> Member Graham, you're recognized.
>> This is the first uh I said as a series of amendments I'll be offering on visual separation.
This report and our investigation revealed that air traffic controllers at DCA were incredibly over reliant on pilot applied visual separation.
The numerous limitations and risk in of see and avoid are well documented in this report and prior NTSB investigations.
If there was ever a time to address this issue, especially from an air traffic control perspective, it would be the result of this in investigation.
I can go on if you want.
>> Uh, no.
Does staff? So, Member Graham, I can appreciate it and I know staff very much would like to support this and I think where we're concerned a little bit is that if controllers are required to issue positive control instructions with every safety alert, could there possibly be some unintended consequences that we're not thinking about? I don't think so, but I'd like to get, if you don't mind, I'd like to get your staff member from the ATC control group back there, get their opinion on this.
>> That would be fine, Mr.
Soer.
So, so yes, we did uh talk about this with uh at some depth yesterday, and our initial our initial feel was the same.
We we liked the idea.
However, when we started to consider the ramifications of potentially requiring the the uh a positive control instruction, that becomes problematic from the standpoint that um there are times a safety alert could be issued without the aid of a radar display, for instance.
And I can't a a controller might not be able to look out there and actually if they're giving this visually just because they see two aircraft coming in close proximity to one another a an instruction that they they can be certain would not be conflicting with something else like they don't know their headings or they don't know exactly the altitude.
So it could actually cause maybe some unintended consequence there.
And then also just the sheer, you know, the number of times if it was every single time they had to that they issued a safety alert, they issued an accompanying control instruction to do something, um, we just felt that there it may introduce unintended consequences.
So, uh, we just after considering all that, the the best thing that we could even, you know, consider is I just Well, I mean that's it.
That's where we landed on it at the end of the day.
>> So, you're not supportive of this anymore.
You helped me write it.
You know that >> it's tough.
I I'm support Well, you're that's why I started that off with that.
I actually was supportive of the concept of it, but after playing it out like that, it made it that it may be very it may be somewhat unrealistic in some respects.
In this case, we certainly think it would have been great.
So I think that uh you know an alternative that I would hope to see is that you know maybe we get um an enhanced emphasis and training on safety alerts at and you know in the FAA to understand what that means to offer an alternate control instruction if feasible and um and and that type of thing to press the issues from that standpoint but we didn't uh we didn't develop anything like that here.
I wish you would have let me know a little earlier, but if that's the case, you're not supportive of I was doing this in in support of your chair's report, group chair report.
>> I understand.
>> And um so I would like to withdraw that re that recommendation.
>> Okay.
>> Which would also have me withdraw my next recommendation, but I would go to the last one then.
>> Okay.
Moving to the last one.
If Okay.
So, please bring up amendment 16.
Am I go to okay to speak on this, Mr.
Soaper? >> Oh, yeah.
You're recognized.
>> Sorry.
Thank you.
I I'll go ahead here.
Uh I move to for a new recommendation the FA update FA job.
Sorry.
Update FA order job order 3120 decimal 4S to ensure controllers receive instructor-led scenario-based initial and annual training on the proper use of tower and pilot applied visual separation via a standalone training module that includes the limitations and responsibilities when applying visual separation.
Second.
>> Yep.
Sorry, I was just reading real quick.
Uh, all right, staff.
>> So, staff uh concurs and I understand we have some language we'd like to add to it if you would consider and I think there's a slide pres prepared.
And I apologize this wasn't floated earlier.
This is a dynamic meeting as we go along.
>> When the chair is finished, I would love to know what ATC staff thinks of this.
Yeah, I would too.
I >> Mr.
Soer, >> can you can you Yeah, and eventually I'm going to want it put back up, but go ahead.
>> So, this one was a result of um obviously we're all supportive of it and sitting back here and going over these actually um Mr.
Hoy brought to my attention that this the wreck as it was written may not land well with the FAA.
They may just come back and say we have this training module already.
Here it is.
We've complied.
And because it didn't ask for something new, didn't maybe outline some other things.
I I believe we we kind of uh wrote that to include everything that was in yours.
I I don't think I missed anything out of that other than the manual.
Uh, and the only reason the manual title wasn't included is because if they do this and they include if they develop the module and include it as refresher as as part of the annual simulator refresher training, it automatically will go into the 3420.
>> So you're concerned as currently drafted that they would just come back and say we already do this.
that that was the concern and our and our point was we want them to not give the same training over again.
We need them to actually teach controllers what the limitations are when applying visual separation.
what their responsibilities remain when they're providing visual separation and things to consider when they are going to apply visual separation like weather, night, day, how far are the aircraft apart, how many other aircraft are you working, do you have an assistant or are you stand alone, are you combined position, all of those things, we want them to teach that and have that guidance in print that they can refer to so hopefully better decisions can be made.
>> Well, this is member Graham's amendment.
I I would be I am willing to remove my recommendation and replace it with and move to replace it with this recommendation.
>> U may I ask for member Graham and staff to consider one change.
Can you put the amendment back up? Well, actually, if you're moving I still need a second first.
>> So, I'll second.
>> There you go.
Thanks, Todd.
I appreciate that.
Thank you.
Um the so you have developed this uh program this new instructor-led scenario-based training and then you say require this training for all new controllers and include on a recurrent basis.
I certainly would want existing controllers to also s receive this training initially.
So if we could just strike new I'd be okay with it.
So all controllers >> or traffic control would agree with that.
>> Oh, even though it says and include on a recurrent basis thereafter and annual.
Okay.
I I I guess I assume that meant all controllers.
But if we need to spell it out more, I'm okay with that cuz this is for you.
I was doing these for you guys cuz I heard you guys from the very beginning that they got issues over there and we need to get it right and we need to do some training.
>> I agree.
And I and and I think that wreck captured all of yours.
I really do.
And and I understand what uh the chairwoman is saying also, which is we need to give it to all controllers now because they haven't been getting the right thing and now continue that throughout.
>> We would be agree with that.
>> I would second that amendment to the amendment.
>> Excellent.
So we know what you want to read it one more time.
Everything that's on the screen minus new All set.
Are we clear? >> Yes.
You want to put it up there and I'll read it, please.
Thank you.
Everything but Okay.
So, my amendment and I move to uh for a new recommendation to read develop a new and comprehensive instructor-led scenario-based training on the proper use of visual separation, both tower and pilot applied.
This training should include information on the inherent limitations of see and avoid, responsibilities when applying visual separation, and guidance for controllers on factors such as current traffic volume, workload, weather, or environmental factors, experience, and staffing that should be considered when applying visual separation.
require this training for all controllers and and include on a recurrent basis thereafter an annual simulator refresher training.
>> I'll second it.
>> Excellent.
Tom, you're up.
You're good.
>> Staff concurs.
>> All right, board.
This is a vote in favor of adding a new recommendation to the FAA as read by member Graham.
Member Graham.
>> I.
>> Member Enman.
>> Hi.
>> Chair Homdy.
>> Hi.
All in favor? >> I would want to add one thing to that.
I think it's important to put up or at least point out for the DCA tower that that requires the simulators to work and that group and organization had been waiting for months and months and months just to have that.
So, I hope it doesn't go without cause with that recommendation.
Just fix the damn machines.
Sorry.
>> Okay.
Uh, excellent.
Totally agree.
Uh, member Ammon, you're recognized.
>> Actually, I'm going to skip past a few of mine.
I think they've been overtaken by a few other events.
And I'm going to go directly, if you will, to amendment seven.
This is a proposed new recommendation to the FAA, evaluate other air traffic control towers to confirm if miles and trail are being adhered to per Trcon and tower agreements.
Second.
>> All right.
Um staff >> staff does not concur.
We have no indication that miles and trail issues are at other airports.
The staff did not does not believe this recommendation is supported based on the findings of this investigation.
>> Member Inman, >> I would agree that we didn't evaluate the rest of the US, but we found a systemic issue that these controllers were constantly asking for miles and trail.
They needed that to do their job safely and so we saw an issue that was not being adhered to at least with their agreement in DCA.
We don't know where else exists in the country and I agree that we didn't evaluate that.
That's the reason why we recommend that the FAA does.
What I don't want to see is again 5 years from now the data is there.
We already knew it and we just didn't want to suggest that the FAA go back and review it.
I know in the helicopter aspect they went ahead and took the recommendations.
They've been surveying other helicopter routes and that was acceptable to us.
But in this case, we do know miles and trail and the spacing was some form of approximate cause to this accident.
Is it happening elsewhere in the country? >> I would add though that the the helicopter routes.
Um we found out the FA had done it across the NAS.
So those were focused on DCA only.
>> There's five of those.
What the routes? >> No.
Five locations in the country that have helicopter routes.
>> I think it's eight.
>> Is it eight? >> But regardless, that's something different.
>> Yeah.
>> Uh >> I was just using that as an example.
I just I think it would be an easy thing for the FAA to do.
And I think there's and I don't know this for for certain, but I would suspect that if there's other towers that are suffering the same way and not able to get their safety recommendations, they would appreciate having another voice in response to this recommendation.
>> May May I thank you.
Uh and may I ask some questions? Uh the agreement between DC Tower, is there a formal agreement between DC Tower and PTOIC Tron on Miles and Trail? Oh, what? There's no there is not.
>> Are there other such tower agreements throughout the national airspace? >> There are others that most most that we see have an agreement with the miles and trail um >> but not all >> imprint.
Correct.
Not all.
>> So, and uh member Enman, I actually I think this is probably workable.
Uh I I'm I'm gonna say my concern with this is actually that if let's say DC Tower had an agreement with PTOIC Tron but the agreement was terrible because they kept asking for uh less you know a a an increase in spacing uh than the Treycon and Tower.
or agreement would actually not ensure safety.
So my concern is actually the words being adhered to per tra and tower agreements that may not ensure safety.
I think your intent is to ensure safety and uh to ensure you're getting adequate spacing.
Um my concern is you might have deficient and safety deficiencies in the applicable TCO and tower agreements.
>> I'm happy to take it.
I'm happy to take a modification to that.
I mean, the idea is the FA's got 30,000 people.
They can't go back and check and take the opinions and give us a report of what they did.
I think that's a lot of people deserve to know that.
Well, I might add too, it's saying what I'm hearing here is it might be good to review that also on an annual basis to make sure it is working for you.
>> Kind of like they're supposed to evaluate the helicopter routes.
>> Yes.
We're going to take five minutes.
Uh we're we're actually going to take make that 15 minutes and I'd like to staff to uh come up and one of us will disappear.
So it does Tom doesn't have a fit.
>> Okay, getting close.
>> We are >> all right.
Member Inman, you're recognized.
I I'm going to actually after consulting with staff and others, I'm going to revise my original recommendation and this will be to the FAA specifically to require each class B or class C ATCT facility to evaluate its existing MIT procedures or agreements to ensure that the spacing provided is appropriate for operational safety and make the results publicly available.
>> I second staff concurs.
>> I just I think again we we have a responsibility beyond just this air traffic control tower to make sure the NAS is safe as possible and I hope this would help achieve it.
>> All right, Mr.
McMurray, you're up.
>> All right, this is for a new funding to the FAA.
Uh member Graham, >> I.
>> Member Enman.
>> Hi.
>> Chair Hammond, >> it's a new recommendation.
>> I'm sorry.
New recommendation I ach.
>> All righty.
So now we get to the uh in recommendations um as amended.
>> Move to adopt.
>> Is there a second? >> Second.
>> All right, Mr.
McMurray.
>> All right, board.
This is a vote in favor of adopting the recommendations as amended.
Member Graham.
>> I.
Member Inman I.
>> Chair Hammond.
>> I.
>> All in favor.
>> All right.
Now we get to uh consider the full report.
Are there any amendments to the report? >> Nope.
All right.
Is there now there are some revisions, right? There are some revisions that will be need need to be made by staff uh to correspond to the new findings and recommendations that were adopted or adjustments to some.
And we recognize that in in uh board order 4D uh it states that uh the board may direct revisions be made to the text uh and uh prepared or finalized by staff and um that if a board member requests that revisions be circulated for review after the board meeting, those will be circulated but for review on uh through the voting process.
process uh giving board members three business days post on those postmeating revisions and so all the board members request that.
So any revisions made would go on PMA uh our voting process so that we can just review those um and uh move that forward.
But uh that means is there a motion? Oh, let me make sure I got it right.
Is there a motion to adopt the final report pending revisions? >> So moved.
Second.
>> All right.
Council.
>> All right.
On adopting the report pending final revisions.
>> As amended.
>> I'm sorry.
>> The report as amended.
>> As amended.
Pending final revisions.
Uh board if in favor say I.
Member Graham.
>> I.
>> Member and I.
Chairman >> I.
>> All in favor? >> All right.
Does any member wish to reserve the right to file a concurring or dissenting statement? Member >> Graham.
>> Sorry.
>> Member Inman.
>> Yes.
>> Me, too.
All three of us.
All right.
I'd like to thank member Graham and member Inman for their preparation in advance of today's meeting and for uh their um our work together during the meeting.
As I stated at the start of this meeting, this was one of the most complex investigations in NTSB's history.
And as we know all too well um you know it was also one of the most significant commercial ava aviation accidents in years.
We all would like to once again extend our deepest sympathies to all the families those who are with us today those who are watching online and weren't able to join us.
We are so truly truly sorry for your loss.
We wish this is not how we met.
I imagine today was very difficult.
This was only the the beginning.
The first step in the process is getting this report out, getting the safety recommendations out, and then we will work together, if you are willing and able, to get our recommendations implemented to see that no other family goes through this again.
We will vigorously advocate for implementation of our recommendations.
And the other day somebody said to me, it was a media person.
Oh, you went on a tirade.
That was the word tirade at that press conference.
You know what? And Todd said this too, we should be angry.
This was 100% preventable.
We've issued recommendations in the past that were applicable here.
We have talked about see and avoid for well over five decades.
It's shameful.
I don't want to be here years from now looking at other families that had to suffer such devastating loss.
We are so sorry.
Action frankly should occur before people die.
The fact is over the last two decades alone, I said we've investigated 163 aviation accidents and 47 incidents resulting from a mid-air collision.
That's over the last two decades.
Midair collision, near midair air collision, or loss of separation.
As a result of those 211 aviation accidents and incidents, 2181 people died.
112 people were injured.
In 94 of those accidents, we raised concerns about see and avoid almost 50%.
the tombstone mentality of how the US government addresses safety improvements in all modes of transit transportation by waiting for people to die to take action by comparing cost versus a and then putting a price tag on people's lives of a whopping 11.6 million per person.
Your loved ones are priceless for all of us here.
You can't put a price tag on life.
At the NTSB, all life is pre precious.
Why do we wait? Why wait for people to die to take action? Why not take measures to prevent tragedy? So, you will remain in our hearts, my hearts.
I will continue.
We will all continue to pray for you every day.
In closing, we all of us up here would like to extend our deep appreciation to all of team NTSB, including our colleagues throughout the office of aviation safety, our investigator in charge, Bryce Banning, the Office of Research and Engineering, the Office of General Counsel, the Office of Safety Recommendations and Communications, and the Office of the Managing Director for their work over the course of this investigation.
and then in the development of an excellent report.
I also want to express our appreciation to the office of the chief financial officer, the office of the chief information officer and the investigative support services division for their vital support of this meeting.
I also want to thank well first of all John O'allahan's family is here.
Are they still here? I don't know.
They might not be.
He did an excellent job on the the visibility study on performance.
That was excellent.
But I want to recognize two people.
Man, they did a lot of work.
Allison Diaz and John Curran.
John, I know your parents are watching.
Sure are.
Um, you did the two of you excellent, excellent job.
Thank you.
Thank you all.
Before we adjourn, I would like to raise a point of personal privilege to thank my special assistant, Lauren Dudley.
This is her last board meeting with us at the NTSB.
I've known Lauren for a really long time.
I worked with her on Capitol Hill.
I know.
And just so you all know, she's not going far.
She's sure to be back in our orbit in her next role as Democratic staff director for the Subcommittee on Aviation under the House Transportation and Infrastructure Committee, where earlier in her career she worked on a variety of aviation issues, including safety and oversight of both the FAA and the NTSB.
Before that, Lauren was investigations and oversight counsel for the TNI committee where she supported the committee's Boeing 737 Max investigation.
Lauren, of course, worked for many years for Eleanor Holmes Norton, which is how I met her.
Lauren joined my team in December 2024 directly from the FAA.
And while it is not in my statement, I'm going to say it was the most memorable of all swearing ins.
We were on the verge of a shutdown.
We did not actually have a shutdown, but I was worried we were going to miss the opportunity to hire her real fast.
So, she was on vacation at Disney.
And over Zoom, I swore her in outside of what roller coaster was that? >> Sorry, it was Universal.
>> The what? >> The Veloca Coaster.
She is a roller coaster fan.
Uh I love it and that was the most memorable.
Um but at FAA she served as assistant administrator for government and industry affairs.
She advised the administrator, the deputy administrator, and other executives on legislative and policy initiatives and helped lead implementation of the FAA reauthorization act of 2024, legislation that also reauthorized the NTSB.
Needless to say, Lauren brought an incredible wealth of expertise to my team and the agency at large.
The tragedy at DCA occurred just weeks into her tenure.
Lauren launched with us and was invaluable on scene and ever since.
Clearly, she is a brilliant policy analyst, attorney, and safety professional with unfailing dedication to our mission.
But what I will personally miss most is her bright spirit, her tremendous grace and poise under pressure, and the warmth she always exudes.
Lauren, you embody somebody I've always thought, which is that safety is about people.
Safety is about people.
You care deeply about the safety of individuals and communities alike.
And your unique background makes you uniquely positioned to catalyze positive change for both and frankly to do everything we want in your new role as d staff director of the subcommittee of aviation.
That was a little bit of a joke, but we'll try to, you know, get her to move our recommendations forward.
But thank you for all you've done and uh for uh for all you have yet to do.
Um I look forward to fighting alongside you and I am very excited that you have this opportunity.
I also served as a staff director of a subcommittee for that committee before I came here to the NTSB.
So actually Lauren, that means you could be an NTSB board member one day, possibly chair.
Anybody else want to? No.
Sure.
Okay.
Uh all right.
We stand adjourned.