[Senate Hearing 112-65]
[From the U.S. Government Publishing Office]



                                                         S. Hrg. 112-65
 
                  AIR TRAFFIC CONTROL SAFETY OVERSIGHT

=======================================================================

                                HEARING

                               before the

       SUBCOMMITTEE ON AVIATION OPERATIONS, SAFETY, AND SECURITY

                                 of the

                         COMMITTEE ON COMMERCE,
                      SCIENCE, AND TRANSPORTATION
                          UNITED STATES SENATE

                      ONE HUNDRED TWELFTH CONGRESS

                             FIRST SESSION

                               __________

                              MAY 24, 2011

                               __________

    Printed for the use of the Committee on Commerce, Science, and 
                             Transportation


                  U.S. GOVERNMENT PRINTING OFFICE
67-894                    WASHINGTON : 2011
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing Office, 
http://bookstore.gpo.gov. For more information, contact the GPO Customer Contact Center, U.S. Government Printing Office. Phone 202ï¿½09512ï¿½091800, or 866ï¿½09512ï¿½091800 (toll-free). E-mail, [email protected].  


       SENATE COMMITTEE ON COMMERCE, SCIENCE, AND TRANSPORTATION

                      ONE HUNDRED TWELFTH CONGRESS

                             FIRST SESSION

            JOHN D. ROCKEFELLER IV, West Virginia, Chairman
DANIEL K. INOUYE, Hawaii             KAY BAILEY HUTCHISON, Texas, 
JOHN F. KERRY, Massachusetts             Ranking
BARBARA BOXER, California            OLYMPIA J. SNOWE, Maine
BILL NELSON, Florida                 JIM DeMINT, South Carolina
MARIA CANTWELL, Washington           JOHN THUNE, South Dakota
FRANK R. LAUTENBERG, New Jersey      ROGER F. WICKER, Mississippi
MARK L. PRYOR, Arkansas              JOHNNY ISAKSON, Georgia
CLAIRE McCASKILL, Missouri           ROY BLUNT, Missouri
AMY KLOBUCHAR, Minnesota             JOHN BOOZMAN, Arkansas
TOM UDALL, New Mexico                PATRICK J. TOOMEY, Pennsylvania
MARK WARNER, Virginia                MARCO RUBIO, Florida
MARK BEGICH, Alaska                  KELLY AYOTTE, New Hampshire
                                     DEAN HELLER, Nevada
                    Ellen L. Doneski, Staff Director
                   James Reid, Deputy Staff Director
                   Bruce H. Andrews, General Counsel
   Brian M. Hendricks, Republican Staff Director and General Counsel
            Todd Bertoson, Republican Deputy Staff Director
                Rebecca Seidel, Republican Chief Counsel
                                 ------                                

       SUBCOMMITTEE ON AVIATION OPERATIONS, SAFETY, AND SECURITY

MARIA CANTWELL, Washington,          JIM DeMINT, South Carolina, 
    Chairman                             Ranking Member
DANIEL K. INOUYE, Hawaii             JOHN THUNE, South Dakota
BARBARA BOXER, California            ROGER F. WICKER, Mississippi
BILL NELSON, Florida                 JOHNNY ISAKSON, Georgia
FRANK R. LAUTENBERG, New Jersey      ROY BLUNT, Missouri
AMY KLOBUCHAR, Minnesota             JOHN BOOZMAN, Arkansas
TOM UDALL, New Mexico                PATRICK J. TOOMEY, Pennsylvania
MARK WARNER, Virginia
MARK BEGICH, Alaska


                            C O N T E N T S

                              ----------                              
                                                                   Page
Hearing held on May 24, 2011.....................................     1
Statement of Senator Cantwell....................................     1
Statement of Senator Hutchison...................................     3
    Prepared statement of Hon. John D. Rockefeller IV, U.S. 
      Senator from West Virginia.................................     3
Statement of Senator Nelson......................................     4
Statement of Senator Rockefeller.................................    36
Statement of Senator Thune.......................................    38
Statement of Senator Lautenberg..................................    40
Statement of Senator Warner......................................    51
Statement of Senator Klobuchar...................................    54

                               Witnesses

Hon. J. Randolph Babbitt, Administrator, Federal Aviation 
  Administration.................................................     5
    Prepared statement...........................................     7
Hon. Calvin L. Scovel III, Inspector General, U.S. Department of 
  Transportation.................................................     9
    Prepared statement...........................................    11
Paul M. Rinaldi, President, National Air Traffic Controllers 
  Association....................................................    19
    Prepared statement...........................................    21
Gregory Belenky, M.D., Research Professor and Director, Sleep and 
  Performance Research Center, Washington State University, 
  Spokane........................................................    30
    Prepared statement...........................................    31

                                Appendix

Response to written questions submitted to Hon. J. Randolph 
  Babbitt by:
    Hon. John D. Rockefeller IV..................................    57
    Hon. Maria Cantwell..........................................    58
    Hon. Tom Udall...............................................    60
    Hon. Mark Warner.............................................    61
    Hon. Kay Bailey Hutchison....................................    62
Response to written questions submitted to Hon. Calvin L. Scovel 
  III by:
    Hon. John D. Rockefeller IV..................................    63
    Hon. Maria Cantwell..........................................    64
    Hon. Tom Udall...............................................    65
    Hon. Kay Bailey Hutchison....................................    65
Response to written questions submitted to Paul Rinaldi by:
    Hon. Maria Cantwell..........................................    66
    Hon. Tom Udall...............................................    67
Response to written questions submitted by Hon. Maria Cantwell to 
  Gregory Belenky, M.D...........................................    68


                  AIR TRAFFIC CONTROL SAFETY OVERSIGHT

                              ----------                              


                         TUESDAY, MAY 24, 2011

                               U.S. Senate,
  Subcommittee on Aviation Operations, Safety, and 
                                          Security,
        Committee on Commerce, Science, and Transportation,
                                                    Washington, DC.
    The Committee met, pursuant to notice, at 2:31 p.m. in room 
SR-253, Russell Senate Office Building, Hon. Maria Cantwell, 
presiding.

           OPENING STATEMENT OF HON. MARIA CANTWELL, 
                  U.S. SENATOR FROM WASHINGTON

    Senator Cantwell. Good afternoon, everyone. Welcome to the 
Senate Committee on Commerce, Science, and Transportation, 
Subcommittee on Aviation Operations, Safety, and Security.
    We're having a hearing today on air traffic control safety 
oversight, and we are joined by witnesses, the Honorable Randy 
Babbitt, FAA Administrator; and the Honorable Calvin Scovel, 
Inspector General, U.S. Department of Transportation; Paul 
Rinaldi, President of the Air Traffic Controllers Association; 
and Dr. Gregory Belenky, Director of the Sleep and Performance 
Research Center at Washington State University. Thank you all 
very much for being here.
    Today, the aviation subcommittee is holding an oversight 
hearing on air traffic control safety. And I know my colleague, 
Senator Thune, will be here soon, but I want to recognize him 
in his new role as Ranking Member for this subcommittee, and I 
say that I look forward to working with him.
    The two issues we are going to focus on basically are the 
advent of a series of recent incidents where air traffic 
controllers fell asleep during nightshifts and the increase in 
the number of reported operational errors by air traffic 
controllers.
    As you know, this year, there have been a number of 
incidents involving air traffic controllers sleeping on duty. 
And I'm deeply concerned, as I know the chairman of the full 
committee is, about these incidents.
    Some are clearly examples of unprofessional behavior on 
part of an individual controller. Their actions are totally 
unacceptable. Controllers do have a professional responsibility 
to come to work rested. Unfortunately, some have used those 
incidents to try and tarnish the reputation of a dedicated 
group of men and women who do work every day to ensure that our 
airspace is the safest in the world.
    Air traffic controllers monitor 35,000 flights daily. Said 
another way, roughly two million air passengers come into 
contact with air traffic control each day. We can talk about 
how the NextGen technologies are going to help us improve this 
system, but we can't forget that at the heart of our air 
traffic control system are approximately 15,000 air traffic 
controllers.
    The incidents do serve to highlight the legitimate safety 
issues of air traffic controller fatigue, particularly those 
working on the midnight shift. There is no escaping the science 
that shift work has the potential to disrupt the circadian 
rhythms of the body and often leads to fatigue.
    Fatigue can seriously impair the work performance of 
individuals, such as air traffic controllers, who perform tasks 
that require consistent concentration. Ultimately, this raises 
concerns for safe operations of the air traffic control system.
    I applaud Secretary LaHood and Administrator Babbitt for 
taking some quick actions. I know that these actions will be 
helpful and hope to improve some of the situation.
    The National Transportation Safety Board has examined and 
made recommendations on air traffic controller fatigue, most 
recently in the aftermath of the 2006 crash of the Comair 5191 
in Lexington, Kentucky.
    It took until 2009 for the FAA and NACTA to get their 
fatigue workgroup underway. My understanding is that they have 
jointly made a dozen recommendations to mitigate air traffic 
controller fatigue.
    The first two recommendations have to do with allowing air 
traffic controllers to recuperate during their break shift, 
particularly in the midnight shift. Historically, the question 
of allowing air traffic controllers to take a break or nap has 
been a political one rather than a scientific one. There are 
decades of science on this issue and we look forward to hearing 
more about it today.
    And I want to say to Dr. Belenky thank you for coming all 
the way from the West Coast, from Washington State University, 
and we look to hear more about the Sleep and Performance 
Research Center and the sciences behind that center.
    I am likewise concerned by the 53-percent increase in 
reported operational errors between Fiscal Year 2009 and 2010. 
Operational errors are situations where planes come too close 
to one another in the air. The number of operational errors 
increased from approximately 1,200 in 2009 to 1,900 last year. 
The errors were of varying degrees. And I'm sure we'll get into 
that during the hearing.
    On March 2, the Committee asked the DOT IG to conduct an 
assessment of the FAA's current categorization of operational 
errors to better understand the impact and actual implications 
of this.
    And last decade, the IG identified the problems with how 
most FAA facilities self reported operational errors. And the 
IG expressed concerns that there was a significant potential 
for underreporting operational errors.
    Beginning in 2008, the FAA made a series of changes. It 
initiated the Air Traffic Safety Action Program, a confidential 
reporting system to encourage air traffic controllers to come 
forward with these reported errors. And it began rolling out an 
automated reporting of operational errors through a new 
software system called the Traffic Analysis and Review Program.
    The Committee is trying to understand if the reasons more 
errors are being reported is because of the FAA finally having 
a more objective and reliable process or whether we are seeing 
just an increase in errors.
    So I thank all of you for being here today. I look forward 
to your testimony at the hearing and coming up with answers on 
how to continue to improve air transportation safety.
    I'd like to call on the Chairman of the full Committee if 
he'd like to make an opening statement.
    The Chairman. I would say to Madam Chair that you said 
everything I was going to say, so I don't see any reason to 
repeat it. So I'll put it in the record.
    [The prepared statement of Senator Rockefeller follows:]

             Prepared Statement of John D. Rockefeller IV, 
                    U.S. Senator from West Virginia

    Good afternoon and thank you, Senator Cantwell. This is Senator 
Cantwell's first hearing as Chairwoman of the Aviation Subcommittee. 
She has picked a timely and important topic.
    Today, we're here to talk about the safety of our air traffic 
control system. In the last 2 months, a series of alarming letdowns by 
controllers have shined a bright light onto a job that usually works 
best when we don't hear anything about it.
    In February, a Knoxville air traffic controller went to sleep while 
working the midnight shift. He made a bed on the floor with couch 
pillows and abandoned his station so he could catch some shuteye.
    A month later, a controller at our national airport just across the 
Potomac River fell asleep on the job. Pilots coming in for a landing 
got radio silence when they contacted the air traffic control tower and 
had to land without that controller's guidance.
    Other incidents of sleeping controllers have since been reported in 
Seattle, Orlando, Cleveland, Miami, Lubbock and Reno.
    Let's be clear on one thing here and now: it's unacceptable for a 
controller to fall asleep on the job. If they do, they should be 
removed immediately. That part is non-negotiable. Someone 5,000 feet in 
the air should never wonder if the controller on the ground has nodded 
off.
    Air traffic controllers have a unique role. They handle runway 
traffic, police the skies and must have eagle-eye attention. I have 
enormous respect for air traffic controllers, most of whom work hard 
and are dedicated, outstanding professionals. We shouldn't tarnish the 
whole profession based on the poor judgment of a few.
    But that's exactly why we are here today. We can't allow recent 
questions about the safety of the FAA to permeate air travel.
    I commend Administrator Babbitt for taking strong action and hope 
the witnesses here today can shed some light on these shortcomings and 
make certain these issues won't happen again.
    I'd like to thank the witnesses for taking the time to be here 
today, and I look forward to your testimony.

    Senator Cantwell. Thank you, Mr. Chairman.
    And the Ranking Member of the full Committee, Senator 
Hutchison.

            STATEMENT OF HON. KAY BAILEY HUTCHISON, 
                    U.S. SENATOR FROM TEXAS

    Senator Hutchison. Well, thank you, Madam Chairman. I think 
you certainly stated the case very well.
    I want to welcome you as the Chairman and John Thune as the 
new Ranking Member of the Subcommittee and look forward to 
working with you, especially on FAA reauthorization, which has 
just been hotlined for the 19th time to be extended.
    And I hope that we can come together in the next month and 
pass a bill that all of us worked very hard to get across the 
floor of the Senate and is now in conference. So this will be a 
major mission for your subcommittee.
    And your topic, today, of course, is very timely, and I 
appreciate all of you being here.
    I want to say I do think that we have had a safe aviation 
safety performance, and, in general, the air traffic 
controllers have done a superb job. We pass 790 million people 
per year through our system, and there are 29,000 to 30,000 
safe flights every day. That is a mark in our favor.
    However, of course, in the last 5 months, we have had 
alarming lapses, and not only the air traffic controllers who 
went to sleep, but, apparently, one was watching a movie during 
the time he was on duty.
    And I think the air traffic control incidents and near-
misses have caused for us to have a hearing. And I think that 
we have to have a system in place, as you must know, that 
catches any kind of weakness in the system and takes action to 
remedy it.
    Mr. Babbitt, you are going to be putting 11,000 new 
controllers in place by the year 2020. There's the turnover, of 
course. So I hope that we will hear that you are going to be 
looking at fatigue factors, training, scheduling, and 
professionalism as we are going into this transition.
    Madam Chairman, I have to say that I have a 3 o'clock 
introduction of a Federal judge candidate, so I'm not going to 
be able to stay for the whole hearing, but I will certainly 
look at the record and be very interested in the results. Thank 
you very much.
    Senator Cantwell. Thank you, Senator Hutchison, and thank 
you for your leadership on trying to move, along with the 
Chairman, the bill through the process.
    Senator Nelson, would you like to make an opening 
statement?

                STATEMENT OF HON. BILL NELSON, 
                   U.S. SENATOR FROM FLORIDA

    Senator Nelson. Madam Chairman, an extraordinary number of 
air traffic controllers do an extraordinary job under 
exceptional circumstances. But the subject of today's hearing, 
I think, underscores all the more why we need to move to the 
next generation of air traffic control.
    We are operating off of a series, a constellation of 
satellites. There will be in the cockpit updated information 
for the crew to know situational awareness at all times, in 
addition to what they're being told from the controllers on the 
ground. And yet we keep dithering and not funding the steps 
that we should toward the next generation.
    It has happened in a lot of our states. Just in April, we 
had a controller asleep in Miami. In March, we had two 
controllers that vectored a Southwest Airlines very close to a 
private aircraft. The next generation of air traffic control 
would help that situation, but, in the meantime, we've got a 
problem that we have to address.
    By the way, if you can figure out fatigue and sleep on air 
traffic controllers, it could sure apply to a lot of other 
professions as well. So I look forward to it. Thank you, Madam 
Chair.
    Senator Cantwell. Thank you, Senator Nelson, and I'm sure 
that you do have a very unique perspective on this and we look 
forward to your questions at the appropriate time.
    Mr. Babbitt, we'll start with you. Thank you for being here 
today and thank you for your testimony, and thank you for your 
leadership during this period of time.

 STATEMENT OF HON. J. RANDOLPH BABBITT, ADMINISTRATOR, FEDERAL 
                    AVIATION ADMINISTRATION

    Mr. Babbitt. Well, thank you very much, and good afternoon 
to you, Madam Chair, Ranking Member Thune, members of the 
Subcommittee, full Committee, Chairman Rockefeller and Ranking 
Member Hutchison. Thank you very much for the opportunity to be 
here to discuss the issues facing the Federal Aviation 
Administration.
    I know that today's hearing will focus on the safety of our 
air traffic control system, and I know that I'll probably get 
some tough questions from you through the course of this 
hearing about some recent incidents.
    But I welcome the opportunity to assure you and assure the 
traveling public that we remain the safest and the most 
efficient transportation system in the world, and to let you 
know that we're also taking a substantial number of actions to 
improve the level of safety.
    Before I address these actions, I think I would be remiss 
if I were to appear before you and not mention the need for a 
multiyear reauthorization. We have a tremendous responsibility 
to enhance the safety of our airspace system and transform it 
from the radar-based system of the last century to the 
satellite-based system of tomorrow. To accomplish our goals, 
the FAA needs a multiyear reauthorization with sufficient 
funding levels.
    As you know, the FAA has not had a steady source of funding 
for over three-and-a-half years now. Instead, we've relied on 
18 short-term extensions of our spending authority. So I'm very 
pleased that both the House and the Senate have passed 
reauthorization bills. We very much appreciate your support. 
It's an important step forward.
    However, if the authorized funding levels that were in the 
House bill--and they are well below what the president proposed 
in his 2012 budget--if the House levels were appropriated, it 
would degrade the safe and efficient movement of air traffic.
    If we delay today's infrastructure investments, the long-
term cost to our Nation, to our passengers, and to our 
environment will far exceed the cost of going forward with the 
technology and the infrastructure improvements we need now.
    I would like to turn now to the reason for today's hearing 
and update you on the actions that we have taken regarding 
fatigue and incidents with air traffic controllers.
    Last month, I had the pleasure of traveling with Paul 
Rinaldi, the President of the National Air Traffic Controllers 
Association, NATCA, to air traffic control facilities all 
around the country. We were on this tour for a call to action, 
promoting both safety and professionalism among the 
controllers.
    The visits reinforced for me that we have a workforce that 
is committed to safety 24/7, but the incidents of employees 
falling asleep on position showed that we have to make changes, 
and we have. We've made significant changes to long-time 
scheduling practices to reduce the possibility of fatigue, 
including establishing a minimum of 9 hours between shifts, and 
we'll do more.
    We've added a second controller on midnight shifts, where 
appropriate, in facilities where there was only one controller. 
We've also changed management in critical positions to ensure 
that we have the right people in the right places.
    We've also, unfortunately, found that it was necessary to 
terminate three controllers who were asleep on the job. This 
type of behavior is completely unacceptable.
    The FAA and NATCA, along with outside experts, have joined 
together to create 12 recommendations regarding fatigue. We've 
now entered into formal discussions with NATCA on these 
recommendations.
    I also want to address your concerns today regarding the 
rise in reported operational errors that we've seen over the 
last few years. I share your concerns. Everyone at the FAA is 
personally committed to the safety of our aviation system. Any 
potential upward trend in errors is deeply troubling.
    However, we believe that this trend largely reflects the 
changes that we've instituted in recent years that encourage 
the reporting of errors. We're gathering more information than 
we ever had previously, and that data will allow us to make 
more informed decisions moving forward to overall enhance the 
safety of the system.
    Our voluntary reporting program is called ATSAP and we 
encourage air traffic controllers to report operational errors 
in exchange for the agency addressing the errors in a non-
punitive manner. This is a program similar to one that exists 
throughout the airline industry.
    These reports have given us information about everything 
from windows that are fogging up in towers to problems with 
radar equipment and ground markings. In Albuquerque, it showed 
us that pilots were missing a new hold-short line on the 
runway. An ATSAP report fixed the problem and it became a 
solution instead of an incident.
    While the incidents at ATSAP were not counted as 
operational errors, I fundamentally believe that this program 
has helped us create a culture today of reporting within the 
FAA. And this is ultimately a very positive change that'll 
enhance safety by enabling us to identify risk and to spot 
trends.
    In addition to this cultural transformation, we've rolled 
out new software that automatically detects operational errors 
and then reports them directly to the FAA's quality assurance 
program for analysis.
    Nobody likes to see operational errors, especially me, but 
we are getting the data today that we need in order to improve 
safety. The American public trusts us to perform our jobs and 
make safety the highest priority every day, year in, year out. 
We're committed to making whatever changes are necessary to 
preserve the trust and to continue to provide the safest and 
most efficient air transportation in the world.
    That concludes my opening statement and I'd be happy to 
answer questions when that time arises. Thank you.
    [The prepared statement of Mr. Babbitt follows:]

    Prepared Statement of Hon. J. Randolph Babbitt, Administrator, 
                    Federal Aviation Administration

    Chairwoman Cantwell, Senator Thune, members of the Subcommittee:
    Thank you for the opportunity to appear before you today to discuss 
the issues facing the Federal Aviation Administration's (FAA) air 
traffic control safety oversight. Several recent incidents and reports 
have called into question the safety of our Nation's airspace and the 
professionalism of our air traffic controllers. Obviously, as 
Administrator, the fact that these incidents occurred and that these 
questions are being asked is extremely disturbing. Today I will 
describe the actions that we have taken to address the areas of 
concern. I want every member of this committee to understand how 
committed Secretary LaHood and I are to working with National Air 
Traffic Controllers Association (NATCA) and our controllers to ensure 
the safety of the system. I believe our Nation's air traffic 
controllers are dedicated and professional and a key reason why we have 
the safest aviation system in the world. But we can always improve, and 
therefore cannot tolerate lapses in judgment when it comes to safety.
    In recent weeks, I have been traveling across the country with 
senior FAA leadership and Paul Rinaldi, the President of NATCA, along 
with his leadership team, on a Call to Action on Air Traffic Control 
Safety and Professionalism. The FAA's safety mandate is a tremendous 
responsibility and air traffic controllers are on the front lines of 
that mandate, day in and day out. We oversee the safe transportation of 
nearly two million people per day. That is why recent events have been 
so troubling. I have been very direct in the conversations I have been 
having with the FAA's workforce. Any incident that calls into question 
the professionalism of air traffic controllers cannot and will not be 
tolerated.
    Together with NATCA, I have communicated that, even though we do 
the right thing over 99.9 percent of the time, we have to do better. We 
cannot have the flying public believe, even for an instant, that they 
cannot trust the men and women who are responsible for getting them to 
their destination safely. So I am asking the workforce to rededicate 
ourselves to the concept of professionalism. I am calling on all 
employees to be responsible, not only for our own actions, but for 
helping to ensure that our colleagues are also committed to excellence. 
I want to create a safety culture that makes it imperative to report 
and correct any potentially unsafe condition or action.
    I am happy to report that we are working hand in hand with NATCA in 
our efforts. We both recognize that air traffic controllers have 
traditionally enjoyed a great deal of respect and admiration, and we do 
not want to see that perception of their profession tarnished. NATCA's 
leadership is willing to work hard with us to demonstrate a united 
front in demanding a new level of excellence. I am proud that FAA's 
relationship with NATCA has improved to the point where this joint 
effort is possible. A few years ago, it might not have been. I think we 
can all agree that working together toward a goal achieves a better 
result than working at odds.
    As this Committee knows, I have been working with the aviation 
industry since shortly after I became Administrator on the concept of 
professionalism, and I think we have made some progress in making it a 
priority. It only makes sense to extend this conversation to the 
controller workforce. What do I think professionalism means? It means 
doing the right thing all of the time, even when no one is looking. It 
means following procedures and ensuring compliance with safety 
standards. It means looking out for each other and making sure that you 
correct colleagues who are not upholding these standards. The business 
of air traffic control is a tremendous responsibility, and I know that 
the controllers feel that responsibility. That is why they also need to 
feel that they are supported.
    This means, on the management side, that we have a responsibility 
to address the areas of risk that have been identified. For example, we 
are looking at how to deal with fatigue, which as this Committee knows 
is a particularly difficult issue. Part of it is staffing, part of it 
is scheduling, part of it is education and, yes, part of it is 
professional responsibility. FAA has been focused on mitigating 
controller fatigue since well before the recently reported incidents. 
FAA and NATCA conducted a joint, in-depth assessment of controller 
fatigue, risks and mitigations beginning in the fall of 2010. Twelve 
recommendations are currently under consideration as a result of that 
review. We want to ensure that we fully understand the impact of any 
changes made before we make them.
    Since the reported incidents, there was an immediate agreement to 
allow for more recuperative time between shifts; a minimum of 9 hours 
in between all shifts. In addition, two air traffic controllers are 
required on duty during the midnight shift at 27 control towers across 
the country where only one controller had been scheduled previously, 
including Reagan National Airport here in Washington, D.C. Other 
scheduling changes have been implemented to accommodate this change 
without immediately hiring additional controllers. The FAA Academy will 
expand and update its fatigue management training to help controllers 
recognize, avoid, and combat fatigue. Not all of the changes are 
universally welcomed. But I am convinced that adding an extra layer of 
safety is the right thing to do.
    The science of fatigue management for air traffic controllers is 
still an emerging discipline. There will undoubtedly be continued 
insights about how to mitigate fatigue and improve safety. Our 
challenge is to implement the benefit of new insights while still being 
good stewards of the taxpayer dollar. I look forward to sharing how FAA 
will move forward in this vital effort.
    The recent incidents have come at a time when we have seen an 
overall increase in the reporting of controller operational errors. 
This is a serious and complex issue for the FAA and one I would like to 
take a moment to discuss.
    For many years now, the aviation industry has been collecting data 
provided voluntarily by airline employees that it and the FAA have been 
analyzing. There is universal agreement that having access to safety 
information we would otherwise not know about has allowed us to 
identify trends and better understand the areas of risk that exist in 
the system so that we can focus our collective efforts on minimizing 
those risks. The FAA believes that this approach has already 
contributed to the remarkable decline of commercial aviation accidents; 
a decline of 82 percent since the late 1990s. With that kind of 
recognized success, it only makes sense to look for a way to expand 
this approach to air traffic control.
    In late 2009, the FAA implemented confidential reporting systems 
and incentives for controllers to provide information directly to 
supervisors. We were seeking to achieve the same gains in knowledge and 
awareness of safety conditions in the air traffic control system that 
we did with the airlines. The reporting program we implemented, the Air 
Traffic Safety Action Program (ATSAP), was similar to those applicable 
for airlines. Further, we deployed additional technology to collect 
safety data. It is certainly fair to note that when the airlines 
implemented confidential reporting and improved flight data recording 
systems, the safety data available increased by a factor of 10 or more, 
so there was certainly an expectation that some significant increase in 
data reported with regard to air traffic would result. The important 
thing to remember is that this is data that we want. This is data that 
we need. This is data that will save lives.
    The above noted changes generated over 28,000 confidential safety 
reports made to ATSAP on numerous safety issues. Although ATSAP filings 
do not get counted as operational errors, FAA believes that the 
improved recording systems combined with the overall safety culture 
that ATSAP and other programs are designed to foster, are at least 
partially responsible for the 53 percent increase in the number of 
losses of separation between FY 2009 and FY 2010.
    The majority of the time, errors and other safety reports provide 
the FAA with knowledge critical to identifying and correcting potential 
risk. The more events the FAA is made aware of, whether through digital 
recording programs or voluntary reporting systems, the greater the 
opportunity to resolve the conditions that resulted in those errors. 
The only way to address system risk is to have as much data available 
as possible to identify problem areas, determine root cause and apply 
sustainable correction. We are now poised to tackle the task of 
fundamentally addressing the issues that contribute to operational 
errors and other safety occurrences.
    But voluntary disclosure doesn't necessarily provide everything we 
need, which is why we are also relying on technology to inform us of 
errors that might otherwise not get reported. We have begun using the 
Traffic Analysis Review Program (TARP), a new software tool that will 
automatically detect losses of separation, collect data, and report 
them directly to FAA's quality assurance group for analysis. TARP 
covers the Terminal area, where we have the highest degree of 
congestion. A similar system was implemented in the En Route 
environment several years ago. While we are still discussing the 
implementation of this program with NATCA, we anticipate its use on a 
24/7 basis within this fiscal year.
    An important thing to note is that all operational errors are not 
created equal. Most operational errors are categorized through a system 
that reflects how much of the safety zone was breached. Most errors are 
classified based on severity as A, B, or C, with A being the closest in 
range and C the furthest apart. Errors in the A category are generally 
the most troubling. Other losses are classified as ``Other'' or 
``miscellaneous'' in order to capture those errors where such precise 
measurements are not possible, for example, non-radar, oceanic, 
terrain, procedural or equipment errors.
    The table below is based on FAA data collections on separation 
events since 2007. The large increase in reports filed between the end 
of 2009 and the end of 2010 is concurrent with the implementation of 
voluntary reporting programs and additional electronic data collection.

------------------------------------------------------------------------
            Category               FY 2007   FY 2008   FY 2009   FY 2010
------------------------------------------------------------------------
A                                       34        28        37        43
B                                      256       318       292       400
C                                      557       663       618     1,059
Other                                  193       340       286       385
------------------------------------------------------------------------
  Total                              1,040     1,349     1,233     1,887
------------------------------------------------------------------------

    In 2010, 1,887 errors were reported, of which 443 were classified 
as A or B. To put these numbers in context, there were more than 133 
million Tower, Tracon and En Route air traffic control operations 
during the same time period in 2010. While the data has not been 
subject to a statistical validation or significance test, it appears 
that error rates in the most serious incident categories (A and B) are 
lower than the overall error rate. I think it is fair to say that, 
while any error is troubling and taken very seriously, the numbers 
above suggest that these types of errors are a relatively a rare event.
    So in conclusion, I would like to reiterate two important points. 
First and foremost, the types of controller incidents that have 
reflected poorly on the FAA's dedication to its safety mission are 
being addressed aggressively, and, where possible, collaboratively to 
identify and mitigate risks, whether they stem from scheduling, 
staffing, technology, training or a combination of thereof. Second, I 
am committed to obtaining the most information possible to understand 
how to make the system safer. I take the rise in reported errors very 
seriously, but it is vital for everyone to understand how important 
information is. I know how disconcerting it is for the public to hear 
on the news that there are flaws or risks in the system. But it is 
essential for the public to put those stories into context and 
recognize that the safety record of commercial aviation is not an 
accident--that it is based on the use of critical information, to make 
informed decisions. These two points work hand in hand. Information is 
vital to improve safety, but where information discloses inappropriate 
actions or attitudes, those individuals who cannot meet the standards 
of professionalism and proficiency that FAA demands will be subject to 
retraining or replacement, as appropriate.
    This has been a difficult time for all of us who are dedicated to 
aviation safety. Our commitment is strong and enduring. But I am 
convinced that these challenges give us the opportunity to move forward 
in a positive and productive way. I look forward to working with 
Congress, FAA's workforce, industry and the public to implement 
improved standards that benefit the safety of a system that is both the 
most complex and the safest in the world.
    That concludes my statement. I will be happy to answer your 
questions at this time.

    Senator Cantwell. Thank you, Mr. Babbitt. We'll look 
forward to that opportunity.
    Mr. Scovel, thank you very much. You can start. Press the 
red button there and----

       STATEMENT OF HON. CALVIN L. SCOVEL III, INSPECTOR 
           GENERAL, U.S. DEPARTMENT OF TRANSPORTATION

    Mr. Scovel. Mr. Chairman, Madam Chairman, Ranking Members 
Hutchison and Thune, members of the Subcommittee, thank you for 
inviting me to this timely hearing on FAA's air traffic control 
system. Recent incidents, including several high-profile 
operational errors, underscore the need for improved oversight 
of this system.
    Today, I will discuss longstanding concerns about FAA's 
process for reporting operational errors and managing the 
Nation's controller workforce.
    Over the past decade, we in DOT's Office of Inspector 
General have repeatedly raised concerns about FAA's reliance on 
controllers to report operational errors and FAA's failure to 
uncover reporting inaccuracies. Our audit and investigative 
work has shown that some operational error reports were 
misclassified as nonevents while others were intentionally 
manipulated to cover up errors.
    FAA's recent deployment of TARP, its automated reporting 
system for the terminal environment, should help reduce these 
weaknesses. However, concerns remain about FAA's efforts to 
accurately count the number of operational errors and identify 
troubling trends.
    For example, it's unclear how FAA will use another recently 
implemented tool, the LoSS Index to improve its operational 
error data and assess risks. Without reliable reporting systems 
and processes, FAA's data on operational errors have little 
value.
    Recent FAA data indicate that operational errors have 
increased substantially in the past year. However, FAA 
officials have stated that the increase is likely due to 
improved reporting practices, not to an actual rise in breaches 
of aircraft separation standards.
    We recently initiated two audits to explore these issues in 
depth. While a lack of trend analyses makes it difficult to 
identify and target the root causes of operational errors, 
several unresolved controller workforce challenges may 
contribute to these errors.
    First, NTSB identified controller fatigue as a potential 
contributing factor in several operational errors. However, FAA 
investigations of operational errors do not always develop 
adequate data on controller fatigue.
    Further, our audit of three complex air traffic control 
facilities in the Chicago area determined that minimal rest 
hours between shifts, on-the-job training and scheduled 
overtime may contribute to fatigue.
    In June 2009, we recommended that FAA determine the extent 
to which fatigue could be causing operational errors and to 
identify and address root causes of fatigue. The NTSB and an 
FAA workgroup have also made numerous recommendations to 
minimize sleep debt. To date, these recommendations have not 
been implemented.
    Second, FAA faces challenges in achieving its goal to 
replace retiring controllers with 11,000 new controllers by 
2020, in large part because requirements in its training 
contract were not well defined and the contract costs exceeded 
the first 2 years' estimates by as much as 35 percent.
    Because the costs were so far above estimates, FAA has been 
unable to implement new approaches and programs that were 
expected to improve the quality and timeliness of controller 
training. At the same time, FAA lacked adequate metrics to 
measure the effectiveness of its controller training program 
and to make needed adjustments. In response to our 
recommendation, FAA recently established more complete metrics 
for evaluating its training program and its effectiveness.
    Finally, FAA's controller placement process does not 
adequately consider controllers' knowledge, skills and 
abilities when assigning them to FAA's more than 300 air 
traffic control facilities.
    As we reported in April 2010, FAA assigns new controllers 
to facilities based primarily on their choice and available 
vacancies not on complexity of operations.
    As a result, FAA is assigning new controller candidates to 
some of the Nation's busiest and most complex air traffic 
control facilities with little consideration of whether they 
have the required skill sets to effectively and safely manage 
traffic at those locations.
    More than 20 facilities that FAA deemed critical to NAS 
operations have a significant percentage of their controller 
workforce in training.
    In 2009, we reported that Southern California TRACON faced 
the prospect of having over 100 controllers in training, more 
than 40 percent of its workforce, potentially overwhelming the 
facility's training capacity.
    More recently, we found that Denver terminal radar approach 
control has 43 percent of its workforce in training. And 
LaGuardia air traffic control tower has 39 percent in training.
    We are currently reviewing FAA's plans to provide its 
critical facilities with the appropriate controller staffing, 
training resources and other support and expect to report on 
our results later this year.
    In closing, I want to commend FAA for ramping up its 
efforts to tackle these complex challenges. Clearly, sustained 
commitment will be critical to ensuring an alert, competent and 
certified controller workforce.
    Madam Chairman, this concludes my prepared statement. I'd 
be happy to address any questions you or members of the 
Subcommittee might have.
    [The prepared statement of Mr. Scovel follow:]

  Prepared Statement of Hon. Calvin L. Scovel III, Inspector General, 
                   U.S. Department of Transportation
    Madam Chairman, Ranking Member Thune, and members of the 
Subcommittee,

    Thank you for inviting me to this important hearing on the Federal 
Aviation Administration's (FAA) Air Traffic Control (ATC) system. As 
you know, the U.S. aviation system is one of the safest in the world--
due in part to the dedicated professionals in FAA and throughout the 
aviation industry. However, several recent incidents have raised 
concerns about the safety of the ATC system. These include reports of 
on-duty controllers falling asleep as well as several high-profile 
operational errors, when controllers failed to maintain minimum 
separation distances between aircraft. These incidents are occurring at 
a time when veteran controllers are retiring at unprecedented rates and 
more new controllers are entering the workforce, requiring 
comprehensive training and placement efforts.
    Administrator Babbitt has acted quickly to respond to these 
concerns, including standing up a task force of external and internal 
experts to review controller training, qualifications, and placement. 
The group is tasked with completing their efforts by the fall of 2011. 
However, further steps are needed to address the challenges of managing 
and overseeing the performance of FAA's controller workforce.
    Over the past decade, we have developed a comprehensive portfolio 
of work involving ATC operations and addressing critical safety and 
workforce management issues. My testimony today will focus on four 
areas involving the ATC workforce that we see as key for effectively 
transitioning to the next generation of air traffic control: (1) 
identifying and addressing the causes of operational errors, (2) 
mitigating controller fatigue risks, (3) adequately staffing the 
controller workforce, and (4) training new controllers.
    In summary, while FAA has acted quickly to address many of the 
recent incidents involving the ATC system, FAA has yet to fully 
identify and mitigate risks related to the management and operations of 
its controller workforce. FAA statistics show a recent significant 
increase in operational errors; however, FAA has not yet determined 
whether the increase is a result of better reporting systems or whether 
there are trends that require mitigating actions. As recent media 
reports have shown, fatigue is a significant concern for the controller 
workforce that FAA must address. Our work and that of the National 
Transportation Safety Board (NTSB) has identified a series of factors 
that create an inherent risk for controller fatigue, but FAA has not 
yet fully implemented recommendations for mitigating that risk. FAA is 
also taking action to hire and train nearly 11,000 new controllers 
through Fiscal Year 2020. However, our work shows that FAA's placement 
process does not adequately consider new controllers' knowledge, 
skills, and abilities when assigning them to ATC facilities, and 
expected innovations to improve the quality and timeliness of 
controller training have not been realized. Ensuring a sufficient, 
competent, and well trained controller workforce is critical to the 
safe and efficient operation of the National Airspace System (NAS).

Weaknesses in Reporting Limit FAA's Ability to Identify Trends in 
        Operational Errors
    FAA statistics indicate that operational errors have risen 
significantly over the past year. However, it is not clear whether this 
reported increase is due to more operational errors being committed or 
to improved reporting practices that have allowed FAA to capture a more 
accurate count of those operational errors that have been committed.
    According to FAA data, the number of operational errors by air 
traffic controllers increased by 53 percent--from 1,234 to 1,887--
between Fiscal Years 2009 and 2010. However, FAA officials acknowledge 
that the increase is likely due to improved reporting practices. 
Specifically, FAA states that the introduction of voluntary, non-
punitive safety reporting programs--such as the Air Traffic Safety 
Action Program (ATSAP) \1\--has encouraged controllers to voluntarily 
report operational errors. The reported increase could also be the 
result of FAA's implementation of the Traffic Analysis and Review 
Program (TARP), which automatically identifies when operational errors 
or other losses of separation between aircraft occur at terminal 
facilities. FAA's recent implementation of TARP represents substantial 
progress in addressing reporting weaknesses. If used effectively and 
consistently at all terminal facilities, TARP could be a significant 
tool for identifying trends in operational errors and addressing 
contributing factors.
---------------------------------------------------------------------------
    \1\ ATSAP is intended to better capture the actual number of 
operational errors and identify and address their root causes.
---------------------------------------------------------------------------
    Historically, FAA's oversight of operational error self-reporting 
has been problematic. Since 2000, our work on operational errors has 
repeatedly raised concerns that nearly 300 FAA terminal facilities 
relied solely on controllers to self report errors. In some cases, we 
found that the self-reporting process was subject to intentional 
manipulation. For example, in both 2005 and again in 2008, our 
investigations at the Dallas/Fort Worth Terminal Radar Approach Control 
(DFW TRACON) facility found that air traffic managers at the TRACON 
intentionally misclassified operational errors as either pilot 
deviations or ``non-events'' 2 to reduce the number of operational 
errors reported at that location. Our 2008 investigation identified 62 
operational errors and deviations that were either incorrectly reported 
as pilot deviations or misclassified as ``non-events.'' \2\ Further, 
FAA's oversight processes failed to uncover this practice despite FAA's 
prior assurances that it would not allow operational errors to go 
unreported. Our recommendations included expediting the deployment of 
the automated TARP reporting system at DFW TRACON because of the 
facility's pervasive problems with self reporting.
---------------------------------------------------------------------------
    \2\ Non-events are those incidents that facility personnel reviewed 
but determined there was no loss of separation.
---------------------------------------------------------------------------
    Concerns remain about whether FAA is accurately counting the number 
of operational errors and sufficiently identifying the trends that 
contribute to them. For example, it is unclear how ATSAP reports are 
factored into FAA's current counts of operational errors.
    Furthermore, NTSB has raised concerns about the reliability of 
FAA's process for assessing and reporting incidents involving the loss 
of separation between aircraft and is currently reviewing airline 
reports of Traffic Collision and Avoidance Systems (TCAS) 
advisories.\3\ Since NTSB issued its final rule requiring aircraft 
operators to report certain TCAS advisories in January 2010, the Board 
has received nearly 950 reports of these collision advisories and has 
initiated investigations into 9 of the more serious incidents.\4\
---------------------------------------------------------------------------
    \3\ An onboard TCAS issues advisories for pilots to take evasive 
actions when the system detects a potential collision with other 
aircraft.
    \4\ After review by NTSB, many of these reports were considered 
``nuisance alerts'' (i.e., situations in which there was no collision 
risk but TCAS generated a resolution advisory). However, about 260 
reports required additional data in order for NTSB to understand and 
evaluate the circumstances that caused the apparent conflict and to 
determine whether further action was warranted.
---------------------------------------------------------------------------
    Further concerns relate to FAA's recent implementation of the new 
System Loss of Standard Separation (LoSS) Index, which is designed to 
capture each incident where aircraft fly closer than separation 
standards permit.\5\ It is unclear how FAA will use LoSS to assess 
operational error risks or improve its error statistics. At the request 
of this Committee and others, we recently initiated two audits to 
assess FAA's implementation and oversight of ATSAP and evaluate FAA's 
process for tracking and reporting loss of separation events and its 
subsequent efforts to analyze and mitigate identified risks through the 
LoSS process.
---------------------------------------------------------------------------
    \5\ The new tool calls for the investigation and analysis of all 
separation losses, not just operational errors. Pilot deviations or 
miscellaneous losses such as emergency descent for pressurization are 
also included. Instances of non-compliance with separation standards 
will be designated as LoSS events.
---------------------------------------------------------------------------
    Clearly, there are a number of questions regarding what is and is 
not reported in FAA's operational error statistics, and we plan to 
answer these questions in our upcoming audits. However, the fact that 
operational errors pose real safety risks is undisputed. FAA needs good 
systems and processes that accurately capture operational errors so 
that the true magnitude of these incidents is known. FAA needs this 
data so it can trend operational errors, identify their root causes, 
and develop actions to effectively address and mitigate them. As we 
progress in our audits into ATSAP and LoSS, we will keep this Committee 
apprised of our findings regarding this critical issue.

FAA Has Not Fully Implemented Recommendations to Identify and 
        Mitigate Fatigue Risks
    Recent reports of controllers falling asleep while on duty 
underscore the need for FAA to take actions to mitigate controller 
fatigue. At the request of Congress, in 2009 we evaluated controller 
fatigue issues at three busy and complex ATC facilities in the Chicago 
area and identified a number of factors that could create potential 
fatigue conditions for controllers. These factors included minimal 
hours between shifts for rest and counter-rotational shifts with 
progressively earlier start times, on-the-job training (OJT), and 
scheduled overtime. We also found that FAA does not consistently 
include fatigue issues as part of its normal operational error 
investigatory process, even though NTSB has identified fatigue as a 
potential contributing factor in several operational errors. While our 
review focused on only the three Chicago facilities, it is likely that 
the fatigue factors that we identified exist at other large air traffic 
control facilities throughout the Nation. We have made a number of 
recommendations to address these concerns, but FAA has not yet 
implemented all of them.

Scheduling Practices and OJT May Create Risks for Controller Fatigue
    Our statistical analyses of schedule information and time and 
attendance data identified factors that could create fatigue conditions 
at all three of the Chicago air traffic control facilities we reviewed 
(Chicago O'Hare, Chicago TRACON, and Chicago En Route Center). For 
example, we found that most controllers at two of the three locations 
were scheduled to work at least one shift each week in which their rest 
period between shifts was less than 10 hours.\6\ Controllers typically 
worked a type of schedule commonly referred to as a ``2-2-1 rotation.'' 
While the configuration of the 2-2-1 rotation may vary, this particular 
scheduling practice usually consists of a work week with two 
consecutive evening shifts, followed by two consecutive day shifts, 
followed by one midnight shift (see table 1).
---------------------------------------------------------------------------
    \6\ FAA Order 7210.3 requires at least 8 hours between shifts for 
rest. For the purpose of our review, we considered a quickturn to be 
less than 10 hours between shifts because FAA was planning on amending 
FAA Order 7210.3 to increase the time available for rest from 8 hours 
to 10 hours.

    Table 1. Example of a 2-2-1 Schedule Rotation Before FAA's Recent
                                 Changes
------------------------------------------------------------------------
       Day              Shift            Start Time         End Time
------------------------------------------------------------------------
1                 Evening            4 p.m.             Midnight
2                 Evening            2 p.m.             10 p.m.*
3                 Day                7 a.m.             3 p.m.
4                 Day                6 a.m.             2 p.m.*
5                 Midnight           10 p.m.            6 a.m.
------------------------------------------------------------------------
*Rest periods between shifts close to FAA minimum requirements.

    Most controllers had at least one ``quickturn'' during the week, a 
schedule characterized by shifts with minimum rest periods between 
them. In addition, we found that none of the three locations had 
established procedures for rotating controllers through more complex 
facility positions during scheduled shifts, even though the complexity 
of these positions can vary extensively.
    We also found that certified controllers at all three facilities 
conducted OJT on a regular basis, which requires a high level of 
concentration and focus on the part of the veteran controller. The time 
spent conducting OJT in our samples ranged from 1 to 5 days per week. 
ATC managers at all three facilities cautioned that OJT is expected to 
increase significantly over the next several years as more trainees are 
added to the workforce.
    We made a series of recommendations for mitigating potential 
fatigue, including amending FAA ATC orders to: (1) increase rest time 
between shifts from 8 hours to 10 hours, (2) increase the time 
available for rest after working a midnight shift, and (3) allow 
controllers to rest when not controlling traffic. FAA agreed with our 
recommendations but subsequently formed a workgroup with the National 
Air Traffic Controllers Association (NATCA) to further review 
controller fatigue issues. The workgroup completed its study and 
presented its findings to the Administrator and union president in 
January 2011 along with 12 recommendations. To date none of the 
recommendations have been implemented, but FAA and NATCA expect to 
finalize their proposed actions later this year.
    NTSB has also made numerous controller safety recommendations 
related to fatigue issues, such as rest periods between shifts, 
scheduling practices, and fatigue awareness training. For example, 
following the 2006 fatal crash of Comair flight 5191, in which NTSB 
examined controller fatigue, NTSB specifically recommended that FAA 
work with NATCA to revise controller work-scheduling policies and 
practices and modify shift rotations to minimize sleep debt and 
decreased cognitive performance. NTSB's recommendations also remain 
open.

FAA Does Not Know the Extent to Which Fatigue Contributes to 
        Operational Errors
    NTSB has identified fatigue as a potential contributing factor in 
several operational errors.\7\ Yet FAA's investigations into the causes 
of operational errors do not consistently address human factors, such 
as fatigue and situational awareness. In our evaluation of controller 
fatigue issues at the three Chicago facilities, we found that their 
operational error investigations did not consistently include a review 
of factors that could cause fatigue. For example, final operational 
error reports that we reviewed at the Chicago En Route Center indicated 
that a controller's work schedule was a ``rotation,'' but there was no 
further information provided to determine the days or the shifts the 
controller actually worked.
---------------------------------------------------------------------------
    \7\ In its April 10, 2007, recommendation letter to FAA and NATCA 
following the crash of Comair flight 5191, NTSB discussed four previous 
air carrier incidents in which fatigue contributed to controller 
errors. Three of these incidents involved runway incursions in Chicago, 
IL, on March 23, 2006; Los Angeles, CA, on August 19, 2004; and 
Seattle, WA, on July 8, 2001. The fourth incident involved a departure 
from a closed runway in Denver, CO, on September 25, 2001.
---------------------------------------------------------------------------
    Accordingly, in our June 2009 report we recommended that FAA 
include potential fatigue factors, such as time off between shifts, as 
a standard part of its operational error investigation process to 
determine the extent that fatigue could be causing these incidents and 
identify actions to address the root cause.\8\ While FAA agreed with 
our recommendation, action has been slow. Last month, in a letter to 
the Chairman of the House Committee on Government Oversight and Reform, 
we identified this recommendation as our most important safety 
recommendation that remains open.\9\ FAA expects to fully address the 
issue next month.
---------------------------------------------------------------------------
    \8\ OIG Report Number AV-2009-065, ``Air Traffic Control: Potential 
Fatigue Factors,'' June 29, 2009. OIG reports are available on our 
website at http://www.oig.dot.gov.
    \9\ Correspondence number CC-2011-024. ``Letter to Chairman Issa on 
OIG's Open Audit Recommendations,'' April 29, 2011.
---------------------------------------------------------------------------
Past FAA Requirements for Staffing Midnight Shifts Were Not 
        Consistently Followed
    Recent media coverage of controllers who fell asleep while on duty 
has drawn attention to the fact that some air traffic control 
facilities were staffed with only one controller during midnight 
shifts. Following the 2006 fatal crash of Comair flight 5191, similar 
concerns were raised regarding single staffed midnight shifts when FAA 
policies issued in 2005 required that two controllers be present in 
towers that provide both tower control and radar services. At the 
request of the then Ranking Member of the House Committee of 
Transportation and Infrastructure and the then Ranking Member of the 
House Subcommittee on Aviation, we reviewed FAA policies that 
prohibited one controller from performing both radar and tower 
controller duties during midnight shifts and determined the extent to 
which the towers covered by the policies complied with them. We 
reported in 2007 that the policies were not being followed 
consistently.\10\ Based on a sample of midnight shifts, we were able to 
statistically project that approximately 11.1 percent of the total 
midnight shifts included in our review period were staffed with only 
one controller.\11\
---------------------------------------------------------------------------
    \10\ OIG Report Number AV-2007-038, ``Review of Staffing at FAA's 
Combined Radar Approach Control and Tower With Radar Facilities,'' 
March 16, 2007.
    \11\ Our review of 20 randomly selected weeks of staffing data for 
midnight shifts at 15 of the 62 facilities in our universe (a total of 
2,100 shifts) identified 234 shifts where only 1 controller was 
scheduled on the midnight shift. Based on the results of our sample, we 
can statistically project (with a 95-percent confidence level) that 
approximately 2,563 or 11.1 percent of the 23,002 total midnight shifts 
(at the 62 facilities in our universe) were staffed with only 1 
controller between August 28, 2005, and September 2, 2006.
---------------------------------------------------------------------------
    More importantly, we found evidence suggesting that the radar and 
ground control duties were combined for substantial periods of time 
even when there were at least two controllers on duty. For example, at 
several facilities, position logs we reviewed showed that all positions 
on midnight shifts were routinely combined and the two controllers on 
duty alternated between working the one position and taking breaks. In 
response to recent events, Administrator Babbitt recently stated that 
FAA will place two controllers on midnight shifts at 27 control towers 
not covered by the 2005 policy. As part of these actions, FAA needs to 
implement corresponding controls identifying when both controllers are 
expected to be on position.

FAA Faces Management Challenges in Training its Controller Workforce
    FAA is taking action to hire and train nearly 11,000 new 
controllers through Fiscal Year 2020 to replace the large numbers of 
retiring controllers hired after the 1981 ATC strike. However, training 
and certifying new controllers have been a challenge, in large part 
because FAA's initial controller training requirements in its training 
contract were not well defined and the contract costs far exceeded the 
first 2 years' estimates. Because costs for basic training needs were 
so far above estimates, current training methods have remained 
essentially unchanged and FAA has not been able to implement new 
approaches and pilot programs expected to improve the quality and 
timeliness of controller training. In addition, FAA's metrics for 
managing its controller training program do not provide a true picture 
of the effectiveness of its training efforts.

Expected Innovations in Facility Controller Training Have Not Been 
        Realized
    Training new controllers to the Certified Professional Controller 
(CPC) level is important for two reasons: (1) only CPCs are qualified 
to control traffic at all positions of their assigned area, and (2) 
only CPCs certified for at least 6 months at their assigned location 
can become OJT instructors for other new controllers. Total training 
can take up to 3 years, and facility training is the lengthiest and 
most expensive part of new controller training. In 2008, FAA awarded a 
contract to Raytheon to administer the Air Traffic Controller Optimum 
Training Solution Program (ATCOTS), a critical component of FAA's plans 
to hire and train 11,000 new controllers by 2020. In designing and 
executing the ATCOTS program, however, FAA did not fully consider the 
number of controllers that needed training under the contract. For 
example, the contract solicitation stated that bidders were expected to 
train approximately 4,000 developmental controllers. However, Raytheon 
estimated that about 5,620 controllers needed training--41 percent more 
than FAA originally estimated. As a result, FAA now faces significant 
challenges in training a new generation of controllers to replace those 
who are retiring.
    As we reported last September,\12\ ATCOTS contract costs and fees 
to date exceeded baseline estimates by 35 percent during the first year 
of the contract (from $81 million to $109 million) and increased by 20 
percent during the second year (from $91 million to $109 million). The 
impact of these cost overruns is that funds have only been sufficient 
to support existing training methods and procedures, not innovative 
training programs.
---------------------------------------------------------------------------
    \12\ OIG Report Number AV 2010-126, ``FAA's Air Traffic Controller 
Optimum Training Solution Program: Sound Contract Management Practices 
Are Needed To Achieve Program Outcomes,'' September 30, 2010.
---------------------------------------------------------------------------
    FAA is taking action to address many of the issues identified 
during our audit. For example, FAA has added a new planning tool for 
evaluating the level of instructor staffing at air traffic facilities. 
FAA is also establishing training priorities to ensure that costs 
remain within baseline estimates. However, it will be difficult for FAA 
to achieve ATCOTS's original training goals or implement any training 
innovations without significantly modifying the existing contract.

FAA Metrics Do Not Provide a Complete Picture of the Effectiveness of 
        Its Training Program
    Accurately assessing the controller training program is critical 
for ensuring a sufficient number of new hires are prepared to replace 
retiring veteran controllers and are assigned to the appropriate level 
and type of facility. Such assessments can also alert FAA to weaknesses 
in its training program that need to be addressed. However, as we 
recently reported,\13\ FAA's metrics for measuring the effectiveness of 
the controller training program are inadequate to identify weaknesses 
and make appropriate and timely adjustments to the program. For 
example, for Fiscal Year 2009, FAA reported a program attrition rate of 
9 percent. However, as Figure 1 shows below, the success rate was only 
4 percent while 87 percent of the controllers were still completing 
their initial training, which can take 2 to 3 years.
---------------------------------------------------------------------------
    \13\ OIG Report Number AV-2011-072, ``FAA Must Improve Its 
Controller Training Metrics To Help Identify Program Needs,'' March 30, 
2011.
---------------------------------------------------------------------------
    Figure 1. FAA's New Controller Training Data for Class Hired in FY 
2009



    When we assessed the number of controllers who successfully 
completed training during a given period of time against those who did 
not, we found a significantly higher attrition rate. For example, we 
grouped the controllers by the Fiscal Year they ended training and then 
identified whether they ended the training successfully or 
unsuccessfully. Our analysis showed that the attrition rate for the 
controllers who ended their initial training in Fiscal Year 2009 was 21 
percent and their success rate was 79 percent.
    We recommended steps FAA should take to measure and present a more 
complete picture of the effectiveness of its air traffic controller 
training program. FAA agreed and is now using more complete metrics for 
evaluating its training successes.

Controller Staffing and Placement Can Be Improved
    FAA's placement process does not adequately consider new 
controllers' knowledge, skills, and abilities when assigning them to 
FAA's more than 300 ATC facilities, which vary extensively in the 
number and complexity of operations. In addition, the recent surge of 
newly hired controllers means there are fewer certified controllers in 
the workforce to control air traffic and provide OJT for new 
controllers. At some critical locations, the percentage of new 
controllers in training is extremely high, which could impact 
operations not only at that location but potentially throughout the 
NAS.

FAA Does Not Adequately Consider Aptitude When Placing New Controllers
    FAA has streamlined its hiring process, and over the past several 
years successfully met its hiring goals for new controllers. However, 
FAA's process for placing new controllers once they are hired does not 
sufficiently evaluate their aptitude before assigning them to complex 
facilities. As we reported in April of last year,\14\ FAA does not use 
results of the Air Traffic Selection and Training Test (AT-SAT) to 
match new controllers' aptitude to the level of facility.\15\ Instead, 
FAA assigns new controllers to locations based primarily on their 
facility choice and available vacancies. As a result, new controller 
candidates are being assigned to some of the busiest air traffic 
control facilities in the Nation with little consideration of whether 
they have the knowledge, skills, and abilities necessary to become 
certified controllers at those locations. We recommended that FAA place 
new controllers based in part on their performance at the FAA Academy. 
FAA partially agreed with our recommendation and initiated a study, 
which the Agency expects to complete by December 2012.
---------------------------------------------------------------------------
    \14\ OIG Report Number AV-2010-049, ``Review of Screening, 
Placement, and Initial Training of Newly Hired Air Traffic 
Controllers,'' April 1, 2010.
    \15\ Air traffic control facilities are categorized by levels (4 
through 12) based on the complexity and number of operations. Level 4 
facilities are the least complex, while Level 12 are the most complex.
---------------------------------------------------------------------------
Critical Facilities May Need More Certified Professional Controllers To 
        Maintain Continuity of Operations
    The increase in hiring has changed the makeup of the controller 
workforce. Currently, new controllers comprise up to 25 percent of the 
ATC workforce compared to 15 percent in 2004. However, this percentage 
can vary extensively by location. For example, Seattle TRACON has 46 
percent of its controller workforce in training, while St. Louis TRACON 
has no controllers in training. Our work at three facilities in 
California (LAX, Southern California TRACON (SCT), and Northern 
California TRACON) \16\ showed that FAA needs to take additional 
measures to ensure that these critical locations have enough certified 
controllers to ensure continuity of safe operations. For example, SCT 
had the highest percentage of existing and planned new controllers of 
the three facilities and had experienced a sharp decline in CPCs over 
the past 5 years. A significant issue was that SCT expected to have 
more than 100 controllers in training--more than 40 percent of its 
workforce--which could overwhelm the facility training capacity. We 
identified four specific focus areas that FAA needed to address: (1) 
making these locations a top priority in FAA's ongoing efforts to 
validate staffing ranges, (2) expanding the use of relocation and 
retention incentives, (3) providing enough instructors and other 
training resources, and (4) ensuring appropriate use of overtime hours.
---------------------------------------------------------------------------
    \16\ OIG Report Number AV-2009-047, ``Controller Staffing at Key 
California Air Traffic Control Facilities,'' April 23, 2009.
---------------------------------------------------------------------------
    Based on our results at Southern California, we initiated a review 
of staffing at other critical NAS facilities. We identified more than 
20 facilities that, if operations had to be curtailed due to a lack of 
certified controllers, could impact the entire NAS. FAA agreed that 
these facilities are critical. Some of these facilities currently have 
a significant percentage of their workforce in training or eligible to 
retire. For example, the Denver TRACON has 43 percent of its workforce 
in training, and LaGuardia ATC Tower has 39 percent. We are reviewing 
FAA's plans to provide its critical facilities with appropriate 
controller staffing, training resources, and other support necessary to 
ensure continuity of facility operations. We expect to report on our 
results later this year.

Conclusion
    While FAA's recent actions to improve ATC operations are steps in 
the right direction, sustained oversight and commitment are needed to 
identify the root causes of ATC incidents and address longstanding 
concerns. Until FAA takes action to develop comprehensive data (such as 
accurately capturing all operational errors), conduct astute trend 
analyses, and develop timely action plans to address controller 
workforce risks and vulnerabilities, FAA cannot ensure it has a 
sufficient number alert, competent, and certified controllers needed to 
effectively manage the challenges of the next generation of air traffic 
control.
    Madam Chairman, this concludes my prepared statement. I would be 
happy to address any questions you or other Members of the Subcommittee 
may have.

 Exhibit A. Significant OIG Air Traffic Control Recommendations and FAA
                        Actions Taken in Response
------------------------------------------------------------------------
                                                FAA's Actions Taken in
      Date            OIG Recommendation               Response
------------------------------------------------------------------------
December 2001     Develop a strategy, in      FAA completed an A-76
                   conjunction with OASIS      study and contracted out
                   deployment, to              its Flight Service
                   consolidate the 61          Stations in 2005 at an
                   existing Automated Flight   estimated savings of $1.7
                   Service Stations.           billion.
------------------------------------------------------------------------
September 2003    Establish milestones for    FAA developed the national
                   completing a national       database for controlling
                   database on all MOUs        MOUs at the national
                                               level.
------------------------------------------------------------------------
June 2004         Compile national            FAA established the
                   statistics and establish    National Training Data
                   a baseline to better        base to manage and track
                   manage the time and costs   controller training at
                   associated with the         the national level.
                   controller OJT process.
------------------------------------------------------------------------
June 2004         Establish a system to       FAA published the 4-year
                   uniformly estimate          attrition estimates by
                   controller attrition by     location in the 2006 and
                   location.                   2007 Controller Workforce
                                               Plan.
------------------------------------------------------------------------
June 2004         Develop an assessment       FAA concurred and stated
                   process for identifying a   it was evaluating data
                   new controller's            gathered from AT-SAT
                   potential to certify at a   scores to determine
                   certain facility level      whether this information
                   and use this information    can improve the
                   in placing newly hired      controller placement
                   controllers.                process. FAA has not yet
                                               completed this
                                               evaluation.
------------------------------------------------------------------------
May 2005          Initiate the planned        FAA completed its efforts
                   assessment of the current   to revise the standards
                   staffing standards for      for towers and en route
                   each facility.              facilities in 2007, and
                                               completed revised
                                               standards for TRACON
                                               facilities in 2009.
------------------------------------------------------------------------
February 2007     Include in the Controller   FAA included staffing
                   Workforce Plan (CWP) the    ranges and actual on
                   staffing ranges for each    board numbers for each
                   facility.                   facility in the CWP.
------------------------------------------------------------------------
April 2008        Permanently change DFW      FAA removed the facility
                   TRACON management team      manager and assistant
                   responsible for the         manager and assigned
                   misclassification of        acting managers until
                   operational errors.         permanent replacements
                                               were selected.
------------------------------------------------------------------------
April 2008        Expedite the early          FAA accelerated the
                   deployment of TARP at DFW   implementation of TARP to
                   TRACON from its current     the end of FY 2008.
                   date of 2011.
------------------------------------------------------------------------
June 2008         Include in the CWP the      Beginning in 2009, FAA
                   actual number of CPCs,      listed the composition of
                   CPC-ITs, and                the controller workforce
                   developmental controllers   by location.
                   by location.
------------------------------------------------------------------------
June 2008         Designate authority and     FAA delegated authority
                   responsibility for          for facility training to
                   oversight and direction     the Manager for Technical
                   of the facility training    Training and Facilities
                   program at the national     Oversight through Order
                   level.                      3120.4M.
------------------------------------------------------------------------
March 2009        Develop milestones for      FAA plans to completely
                   implementing Traffic        implement TARP by
                   Analysis and Review         September 2011.
                   Program (TARP) as a full-
                   time separation
                   conformance tool
------------------------------------------------------------------------
June 2009         Expand operational error    The next version of FAA's
                   investigatory               Air Traffic Safety Action
                   requirements to include     Program submitter report,
                   more detailed information   scheduled for
                   on fatigue factors, such    implementation in summer
                   as overtime, OJT, and       2011, will contain the
                   work schedules.             fatigue data capture
                                               questions.
------------------------------------------------------------------------
September 2010    Ensure that the ATCOTS      FAA estimates that
                   program office has enough   additional personnel will
                   qualified personnel to      be added by December 31,
                   oversee the contractual,    2011.
                   financial, and
                   operational aspects of
                   the program.
------------------------------------------------------------------------


        Exhibit B. OIG Published Reports on ATC Issues since 2001
------------------------------------------------------------------------
  Report Number              Report Title               Date Published
------------------------------------------------------------------------
AV-2002-064       Automated Flight Service Stations:  December 2001
                   Significant Benefits Could be
                   Realized by Consolidating AFSS
                   Sites in Conjunction with
                   Deployment of OASIS
------------------------------------------------------------------------
AV-2003-040       Operational Errors and Runway       April 2003
                   Incursions: Progress Made, but
                   the Number of Incidents is Still
                   High and Presents Serious Safety
                   Risks
------------------------------------------------------------------------
AV-2003-059       FAA's Management of and Control     September 2003
                   Over Memorandums of Understanding
------------------------------------------------------------------------
AV-2004-060       Opportunities To Improve FAA's      June 2004
                   Process For Placing and Training
                   Air Traffic Controllers in Light
                   of Pending Retirements
------------------------------------------------------------------------
AV-2004-085       Audit of Controls Over the          September 2004
                   Reporting of Operational Errors
------------------------------------------------------------------------
AV-2005-060       Controller Staffing: Observations   May 2005
                   on FAA's 10-Year Strategy For The
                   Air Traffic Controller Workforce
------------------------------------------------------------------------
AV-2006-021       FAA Has Opportunities to Reduce     December 2005
                   Academy Training Time and Costs
                   by Increasing Educational
                   Requirements for Newly Hired Air
                   Traffic Controllers
------------------------------------------------------------------------
AV-2006-050       Report on the Air Traffic           June 2006
                   Organization's Management
                   Controls Over Credit Hours
------------------------------------------------------------------------
AV-2007-032       FAA Continues To Make Progress In   February 2007
                   Implementing Its Controller
                   Workforce Plan, But Further
                   Efforts Are Needed In Several Key
                   Areas
------------------------------------------------------------------------
AV-2007-038       Review Of Staffing At FAA's         March 2007
                   Combined Radar Approach Control
                   and Tower With Radar Facilities
------------------------------------------------------------------------
AV-2007-048       Controls Over the Federal Aviation  May 2007
                   Administration's Conversion of
                   Flight Service Stations to
                   Contract Operations
------------------------------------------------------------------------
AV-2007-050       Progress Has Been Made in Reducing  May 2007
                   Runway Incursions, but Recent
                   Incidents Underscore the Need for
                   Further Proactive Efforts
------------------------------------------------------------------------
AV-2008-055       Review of the Air Traffic           June 2008
                   Controller Facility Training
                   Program
------------------------------------------------------------------------
AV-2009-045       FAA's Process for Reporting and     March 2009
                   Investigating Operational Errors
------------------------------------------------------------------------
AV-2009-047       Controller Staffing at Key          April 2009
                   California Air Traffic Control
                   Facilities
------------------------------------------------------------------------
AV-2009-059       Training Failures Among Newly       June 2009
                   Hired Air Traffic Controllers
------------------------------------------------------------------------
AV-2009-065       Air Traffic Control: Potential      June 2009
                   Fatigue Factors
------------------------------------------------------------------------
AV-2010-049       Review of Screening, Placement,     April 2010
                   and Initial Training of Newly
                   Hired Air Traffic Controllers
------------------------------------------------------------------------
AV-2010-071       Review of FAA's Call to Action      July 2010
                   Plan For Runway Safety
------------------------------------------------------------------------
AV-2010-126       FAA's Air Traffic Controller        September 2010
                   Optimum Training Solution
                   Program: Sound Contract
                   Management Practices Are Needed
                   To Achieve Program Outcomes
------------------------------------------------------------------------
AV-2011-072       FAA Must Improve Its Controller     March 2011
                   Training Metrics To Help Identify
                   Program Needs
------------------------------------------------------------------------
------------------------------------------------------------------------
Note: OIG reports are available on our Website at http://
  www.oig.dot.gov.


    Senator Cantwell. Thank you, Mr. Scovel.
    Mr. Rinaldi, welcome to the hearing. Thank you for your 
testimony.

           STATEMENT OF PAUL M. RINALDI, PRESIDENT, 
          NATIONAL AIR TRAFFIC CONTROLLERS ASSOCIATION

    Mr. Rinaldi. Thank you, Chairman Rockefeller and Madam 
Chair Cantwell----
    Senator Cantwell. Is your microphone on? Is it----
    Mr. Rinaldi. Is it on? There we go.
    Chairman Rockefeller, Madam Chair Cantwell, Ranking Member 
Thune, members of the Committee, I'm the President of the 
National Air Traffic Controllers Association, who represents 
over 15,000 air traffic controllers within the FAA.
    Our controllers are dedicated professionals with a passion 
to run the safest, most efficient system in the entire world. 
According to a recent MIT study, you're safer on a commercial 
airline in this country than you are on an escalator.
    Last year, we had over nine million commercial flights with 
zero fatalities. That's something we're very proud of, but we 
can always do better, and we can always make the system safer.
    In this testimony, I would like to address three topics. 
One would be the professionalism of the air traffic control 
system. Two would be the increase in operational errors. And 
three would be fatigue in the air traffic control work 
environment.
    I need to be very clear. The air traffic controllers are 
very professional. We work day in and day out, 24 hours a day, 
7 days a week, 365 days a year to run the safest, most 
efficient system in the world.
    On an average day, we work over 700,000 operations. We save 
lives. We make emergency situations look routine, and that will 
never find its way into the press.
    We are very unfortunate to have these incidents that have 
happened and found its way into the press, and we are not 
satisfied with it. We are not happy with it, and we have worked 
very closely with Administrator Babbitt to ensure that this 
will not happen again. We are proud professionals and dedicated 
to the safety of the flying public.
    I, along with Executive Vice President Trish Gilbert, have 
traveled throughout the country with Administrator Babbitt, 
with Deputy Administrator Michael Huerta, to address these 
issues with the controllers to ensure that professionalism is 
first and foremost in the operation, that the safety of the 
flying public is first and foremost and stays on the focus of 
every air traffic controller in the system.
    Over a year ago, we started working jointly with the FAA to 
develop a professional standards program, which is peer to 
peer, to really instill that we stay focused on the safety of 
the flying public.
    I've heard from statements today there is great interest in 
the increase in operational errors in the air traffic control 
system. I'd like to make two points on that. First, the vast 
majority of operational errors are really not safety risks, 
and, second, we don't believe comparing 2010 numbers to 
previous years is appropriate. We had a big change in the FAA 
and a change for the better.
    I commend Administrator Babbitt for bringing a new culture, 
of just culture of reporting every instance from the lowest to 
the deepest procedural issues, so that we can address every 
safety issue in the system to enhance the safety of the system 
anyway we possibly can.
    Fatigue is real in our work environment. It is something 
that we have tried to work with the agency with the previous 
administrator, but we are working with the administrator, 
Administrator Babbitt, for the last 12 months trying to put 
together 12 recommendations, along with science and NASA 
scientists and the FAA and guidance with the NTSB, to address 
fatigue. It is a high-stress occupation and it is something 
where perfection is the bottom line and anything less than 
perfection is completely unacceptable.
    In closing, NATCA is on the forefront of improving the 
safety of the National Airspace System. We have pushed for 
years for the ATSAP program to voluntarily report situations 
that might cause safety problems in the system. We jointly 
develop professional standards with the FAA, and, over the last 
year, we worked real hard with scientists, with the FAA to come 
up with 12 recommendations to improve fatigue in our work 
environment. We look to implement these 12 recommendations as 
soon as possible. We have to be 100 percent 100 percent of the 
time. Anything less than that is completely unacceptable.
    I can't stress enough that the men and women--the fine men 
and women of the National Air Traffic Controllers Association 
work the safest, most efficient, most complex system in the 
world, and I want to make sure we do focus on that.
    I thank you for your time and I look forward to answering 
any of your questions.
    [The prepared statement of Mr. Rinaldi follows:]

           Prepared Statement of Paul M. Rinaldi, President, 
              National Air Traffic Controllers Association

Introduction
    The National Air Traffic Controllers Association (NATCA) is the 
exclusive representative of over 15,500 air traffic controllers serving 
the Federal Aviation Administration (FAA), the Department of Defense 
(DOD), and the private sector. In addition, NATCA represents the FAA's 
Alaska flight service specialists and approximately 1,200 FAA 
engineers, 600 traffic management coordinators, 500 aircraft 
certification professionals, agency operational support staff, regional 
personnel from FAA's logistics, budget, finance and computer specialist 
divisions, as well as agency occupational health specialists, nurses 
and medical program specialists.
    Air traffic controllers are dedicated to ensuring that our National 
Airspace System (NAS) is the safest in the world. In order to maintain 
that safety, our controllers work to modernize the NAS, promote new 
technology, and improve safety procedures. Controller skills are put to 
work every day as they handle an impressive volume of flights--air 
traffic controllers monitor takeoff and landing for more 70,000 flights 
each day, safely moving nearly two million passengers throughout the 
country. Air traffic controllers handle these flights in complex air 
space with roughly 5,000 planes in the sky at any given moment.
    With about 64 million take-offs and landings each year, our highly 
trained controller workforce ensures safety. According to MIT, flying 
is 22 times safer than driving; and the chance of a fatality on a 
scheduled flight in the U.S. is one out of 14 million.
    Air traffic controllers take considerable pride in their work. The 
controller work ethic and commitment to safety is not reflected in the 
high-profile incidents that recently gained media attention; the 
professional reputation of air traffic controllers should not be 
tarnished by a few incidents.
    In this testimony, we would like to address the three policy 
changes that have led to increased reporting of operational errors. We 
will also discuss the series of incidents that gained widespread media 
attention, specifically explaining what we believe is the root cause of 
these incidents: fatigue. Our joint FAA-NATCA Fatigue Workgroup has 
made 12 recommendations for mitigating the risks associated with the 
midnight shift and fatigue. As subject matter experts qualified to 
determine inherent risks in air traffic control, we are working with 
the FAA to ensure that the risk of fatalities and errors are mitigated 
to their lowest possible levels.

Recent Increase in Reporting of Operational Errors
    As per the Committee's request, NATCA will address the policy 
changes that have led to the increase in reported operational errors in 
the NAS. Just this month, the Department of Transportation Inspector 
General (DOT IG) cited a 53 percent increase in the number of reported 
operational errors between Fiscal Years 2009 and 2010 (from 1,234 to 
1,887).\1\ The increase can be attributed in large part to policy 
changes intended to improve the identification and reporting of 
operational errors and promote a safety culture in which errors can be 
reported without fear of punitive measures, as well as certain strains 
on the system associated with high ratios of trainee to fully certified 
controllers.
---------------------------------------------------------------------------
    \1\ Department of Transportation Inspector General testimony before 
Subcommittee on Transportation, Housing, and Urban Development, and 
Related Agencies, ``The Federal Aviation Administration's Fiscal 2012 
Budget Request: Key Issues Facing the Agency.'' May 12, 2011.
---------------------------------------------------------------------------
The Definition of Operational Errors
    By definition, an operational error is an event that involves a 
loss of separation, and is attributable to an element of the air 
traffic system (see appendix for full technical definition). 
Operational errors are not always near-collisions or potential aircraft 
accidents; in most cases they are breaches of procedure or safety 
buffers that require investigation to determine cause and how to 
prevent recurrence.
    It is important to note that not all operational errors are a 
result of a controller error. An operational error can also be a system 
error, such as an equipment malfunction or an improperly worded 
procedure that leads to a loss of separation.
    Operational errors are categorized by risk associated with each 
event. Where it can be measured in terms of distance, separation losses 
are categorized in range bands designated as: A, B or C operational 
errors, with A being the closest range and C the furthest apart. For 
other separation losses, where such precise measurements are not 
possible, for example, non-radar, oceanic, terrain or procedural 
errors, or in the event of procedural or equipment malfunctions they 
are classified as ``Other'' or ``Miscellaneous'' operational errors.

The New Safety Culture at the FAA--Count Every Error and Learn from It
    The safety culture that NATCA and the FAA have worked to create 
demands that all categories of errors be reported and counted as 
accurately as possible. One highly accurate program identifies errors 
imperceptible to the human eye, and will continue identifying 
increasing numbers of errors once it is fully operational and 
monitoring all air traffic (it is currently only employed for a certain 
number of hours per week). In addition to this precise error 
identification program, in July of 2008, NATCA and the FAA introduced a 
confidential safety reporting system intended to address systemic 
safety concerns rather than treat individual errors punitively, 
creating an atmosphere in which air traffic professionals feel 
confident that reporting errors will not result in punitive measures. 
The goal of each of these programs is to increase reporting of errors 
so they can be utilized to evaluate, propose, and implement changes to 
further the goal of risk mitigation. Simply stated, the best way to 
increase safety is to find every error and use this data to increase 
the safety of the system and, ultimately, the passengers and users of 
the system. While these programs likely account for most of the 
increase in reported errors, another contributing factor is the strain 
placed on the entire system and workforce by the hiring of 7,800 new 
air traffic controllers in the last 5 years.\2\
---------------------------------------------------------------------------
    \2\ FAA, ``A Plan for the Future 10-Year Strategy for the Air 
Traffic Control Workforce 2011-2020.''
---------------------------------------------------------------------------
Increased Accuracy in Error Identification Adds to Increased Number of 
        OEs
   Traffic Analysis and Review Program (TARP) Identifies 
        Minimal Losses of Separation that Cannot Be Identified by the 
        Human Eye. The FAA has started using TARP more and more over 
        the past year. This automated system identifies when 
        operational errors or other losses of separation occur at 
        terminal facilities. It measures down to 1/100th of a mile, 
        measurements that cannot be seen with the human eye, and thus 
        were previously unreported. This can be seen in the fact that 
        category C errors (those with the least significant loss of 
        separation of any errors) have increased from 618 in FY 2009 to 
        1,059 in FY 2010. In layman's terms, this program is picking up 
        errors that have previously existed in the system, but have 
        never been counted as errors. More importantly, TARP will soon 
        be operating continuously, and we expect that this precise 
        electronic monitoring of the operation will actually result in 
        another increase in reporting of errors.

Safety Culture Enables Air Traffic Professionals to Report Errors 
        without Fear of 
        Reprisal
   The Air Traffic Safety Action Program (ATSAP) allows 
        controllers to report errors without fear of reprisal or 
        punitive measures. As part of efforts to enhance the safety 
        culture of air traffic control and meet Congressional mandates, 
        voluntary non-punitive programs have been implemented for the 
        open reporting of safety concerns by controllers and other FAA 
        employees. The result of this has been an environment in which 
        operational errors are openly reported, as never before.

    The Air Traffic Safety Action Program (ATSAP), is modeled after the 
        very successful program used for over 12 years for airline 
        pilots, known as Aviation Safety Action Program (ASAP). Like 
        ASAP for pilots, ATSAP has created an environment in which 
        employees can report mistakes (operational errors, operational 
        deviations, and other reportable events) to management without 
        fear of reprisal from their employer. This has led to an 
        increase in reported errors.\3\ It is important to note that 
        ATSAP does not remove accountability or responsibility from 
        controllers; it takes a more systemic approach to addressing 
        safety issues. ATSAP also improves the FAA's ability to provide 
        additional training because an Event Review Committee (ERC), 
        which reviews and analyzes the ATSAP reports to identify actual 
        or potential safety problems, proposes solutions for those 
        problems. The ERC recommends training with facility input, 
        which provides a more measured approach compared to knee jerk 
        reactions or punitive approach that had been taken in the past.
---------------------------------------------------------------------------
    \3\ Note: ATSAP has contributed to the increased number of reported 
errors. Confidential reports made by front-line employees to ATSAP are 
otherwise unknown to the FAA unless individuals choose to also report 
directly to their first level supervisors. Thanks to the new safety 
culture, they are likely to report errors in both systems. This is the 
identical process successfully in use by the airlines and is producing 
significant amounts of safety data for the Air Traffic Organization 
(ATO).

   Our collaborative efforts with the Air Traffic Organization 
        (ATO) to increase reporting through ATSAP and to address those 
        safety issues that contribute to high-risk events are 
        essential, and NATCA looks forward to working with the ATO to 
        develop and implement meaningful strategies to continue to 
---------------------------------------------------------------------------
        reduce risk in the NAS.

   Elimination of operational error (OE) quotas that prevented 
        managers from reporting all errors. As ATSAP was implemented, 
        the facility operational error limits, or quotas, were removed 
        in an effort to get more realistic and honest reporting. In 
        other words, each facility had a yearly or quarterly maximum 
        number of operational errors that they were expected not to 
        exceed. Prior to the elimination of the operational error 
        quotas, a manager's performance was tied to the number of 
        operational errors as a metric during their performance 
        evaluations. Thus the more reported operational errors charged 
        against a facility, the greater the negative impact on the 
        facility manager's evaluation and pay. By removing this 
        disincentive to report operational errors, we have no doubt 
        seen increased reporting of errors up the chain that were 
        occurring all along but had not been reported. In addition, the 
        rise in reported errors was concurrent with the implementation 
        of ATSAP and removal of the error quotas in August 2008.

The Stress on the System and the Workforce
    As we are all aware, the NAS has recently hired unprecedented 
numbers of new controllers, which is contributing to a strain on the 
system and workforce. High trainee ratios is an issue that NATCA has 
been warning about for over 5 years, and testified about before the 
Senate Subcommittee on Aviation Operations, Safety, and Security in 
March 2007,\4\ and before the House Subcommittee on Aviation in June 
2008 \5\ and March 2007.\6\ Our message was consistent: The recent 
surge in new hires is placing a serious strain on the system and 
leading to safety concerns as experienced controllers retire and are 
replaced with trainees who require several years to become fully 
certified controllers.
---------------------------------------------------------------------------
    \4\ Testimony of Patrick Forrey, President, NATCA Before the Senate 
Committee on Commerce, Science, and Transportation's Subcommittee on 
Aviation Operations, Safety, and Security, March 8, 2007.
    \5\ Testimony of Patrick Forrey, President, NATCA Before the House 
Transportation and Infrastructure Committee Subcommittee on Aviation, 
June 11, 2008.
    \6\ Testimony of Patrick Forrey, President, NATCA Before the House 
Transportation and Infrastructure Committee Subcommittee on Aviation, 
March 22, 2007.

   Large numbers of new hires require additional resources to 
        train. NATCA testified before the House Aviation Subcommittee 
        in May 2007 and again in June 2008 about the strains of hiring 
        thousands of new controllers in a relatively short period 
        (7,800 new hires over the past of 5 years) would have on the 
        ATC system. In the long-term, these new hires will enhance the 
        safety and efficiency of our NAS, but in the short-term, this 
        places a strain on facilities where they train because while 
        achieving certification on position, trainees work under the 
        direction of a fully certified controller or on-the-job-
        training instructor (OJTI). Their OJTI is therefore taken away 
        from his normal controller duties, leaving one more position to 
        staff in his absence. We have seen that fewer controllers in a 
        facility or a higher trainee to controller ratio may also lead 
        to an increased safety risk. Trainees currently account for 22 
        percent of the workforce \7\ across the system. For example, 
        Chicago TRACON (C90) currently has 20 percent trainees, while 
        Atlanta ARTCC (ZTL) has 26 percent trainees working at their 
        facility.\8\
---------------------------------------------------------------------------
    \7\ FAA payroll data, July 2010.
    \8\ FAA, ``A Plan for the Future 10-Year Strategy for the Air 
Traffic Control Workforce 2011-2020.''

   On the job training takes a toll on the instructing 
        controller. Providing on-the-job-training (OJT) to a new hire 
        is extremely demanding, as the OJTI needs to be aware of every 
        transmission and every keystroke the trainee makes. During OJT, 
        a trainee works live air traffic, while the OJTI monitors both 
        the trainee's actions and the radar or runway environment. The 
        OJTI is held responsible for any errors made by the trainee. 
        This essential training process increases workload for the OJTI 
        and contributes to fatigue, particularly when these controllers 
        are expected to train on nearly a daily basis. These 
        instructors may also lose their proficiency while spending the 
        majority of their time training others instead of working on 
        position themselves, removing highly-trained, certified 
        controllers from the operation and exacerbating the staffing 
---------------------------------------------------------------------------
        shortage.

    It is clear that the policy changes implemented by the FAA to 
gather as much operational error data as possible in order to create a 
safer ATC system has led to the increase in reported operational 
errors. The increase can be attributed in large part to policy changes 
intended to improve the identification and reporting of operational 
errors and promote a safety culture in which errors can be reported 
without fear of punitive measures, as well as certain strains on the 
system. It is important to note that in the coming months, increased 
use of programs such as TARP are expected to result in another spike in 
errors--even with partial implementation, we have seen an increase in 
Category C errors from 618 in FY 2009 to 1,059 in FY 2010, largely 
because TARP can identify errors too precise for humans to identify. 
While the human error will always be a part of the system, it is our 
responsibility to work to identify and rectify errors as often as 
possible. To that end, NATCA fully supports and endorses the increased 
safety culture and the use of the ATSAP Program.
    While much of the increase in reported operational errors can be 
attributed to improved reporting accuracy, we also acknowledge that the 
combined stress of high attrition rates from 2006 to 2009 and the 
resulting surge in new hires has left the system overwhelmed with 
trainees. While this is productive in the long run, the essential 
process of training these new hires uses scare resources such as 
controller time and energy, placing an additional burden on the system. 
NATCA strongly encourages the FAA to conduct a full staffing survey to 
find appropriate staffing levels to mitigate this strain.

Recent Incidents and the Effects of Controller Fatigue
    The Committee has requested that we address the incidents that 
occurred over the course of 4 weeks in March and April of 2011. These 
nine incidents gained significant media attention, resulting in 
veritable media frenzy. These incidents involved supervisors and 
controllers who had allegedly fallen asleep while on position during 
the midnight shift (midshift), a controller who had been watching a DVD 
while on position, and one incident in which the First Lady's plane 
experienced a loss of separation. Additionally, some have included an 
event when a jumbo jet clipped a regional jet at JFK, despite the fact 
that this was not a controller error. The FAA is treating each incident 
as unique and investigating each one. To date, they have suspended 
several controllers from working traffic and one is no longer with the 
Agency. The Agency has stated it will continue to investigate each 
incident and take action as appropriate.

NATCA Response To The Incidents
    When the first incident occurred, NATCA responded swiftly and 
firmly, issuing a statement declaring that safety is NATCA's number one 
priority and our members are committed to performing their critical 
function in the safest, most professional manner possible. NATCA does 
not condone sleeping while on position. In letters to Congress and 
other public communication, NATCA President Paul Rinaldi emphasized 
that the professional reputation of air traffic controllers should not 
be tarnished by these incidents. President Rinaldi and FAA 
Administrator Randy Babbitt jointly wrote an editorial in USA Today 
outlining their concerns and steps both the Union and Agency are taking 
to reduce fatigue and safety concerns. (See Appendix for article). Our 
controllers work every day to ensure the safety of the system, and 
NATCA believes that the safety of the system is paramount.
    However, President Rinaldi also quickly pointed out that the 
controller fatigue issue is real and relevant in this discussion, 
especially when addressing issues regarding the midnight shift. Fatigue 
has existed in our system for many years, and NATCA has a consistent 
record of encouraging the Agency to address the issue. NATCA has warned 
about the safety concerns associated with staffing the midnight shift 
with only one controller. We have always insisted that if the FAA 
decides to keep a tower open overnight, they should staff the tower 
with a minimum of two controllers. President Rinaldi praised the FAA's 
announcement of increased staffing as a strong first step in ensuring 
that fatigue is mitigated on midnight shifts.
    NATCA President Rinaldi and Executive Vice President Trish Gilbert 
also joined a Call to Action tour with FAA Administrator J. Randolph 
Babbitt and other FAA senior officials to begin an honest dialogue with 
employees about the recent incidents. They discussed at length the 
largest underlying problem that contributed to the majority of recent 
events: fatigue. The majority of incidents occurred during midnight 
shifts, when fatigue is most problematic. For more than a decade NATCA 
has expressed its deep concerns about increasing controller fatigue. 
Our national constitution calls for the ending of single staffing on 
the midnight shift, and for years we have advocated past 
Administrations and Congress on the need to find more complete 
solutions to controller fatigue before it is too late.

Impact of Fatigue
    NATCA has not been alone in warning about the dangers of fatigue. 
The National Transportation Safety Board (NTSB) and others have joined 
NATCA in issuing these warnings. In April of 2007, in response to the 
August 27, 2006 Lexington crash of Comair Flight 191 (Delta Connections 
Flight 5191) in which 49 people were killed, the NTSB issued parallel 
safety recommendations to both the FAA and NATCA. The recommendations 
urged the Parties to work together to reduce the potential for 
controller fatigue by revising controller work-scheduling policies and 
practices to provide rest periods long enough for controllers to obtain 
sufficient restorative sleep; by modifying shift rotations to minimize 
disrupted sleep patterns; and to develop a fatigue awareness and 
countermeasures training program. The resulting joint Fatigue 
Workgroup, which came into being with the 2009 Collective Bargaining 
Agreement (CBA), was required to develop a fatigue management system; 
to identify and mitigate workplace fatigue inherent in a 24/7 
operation; and to refer recommendations for action.
    The Fatigue Workgroup consisted of FAA managers and NATCA members, 
supported by scientists from seven different components of the FAA, 
including Aerospace Medicine and FAA's Civil Aerospace Medical 
Institute (CAMI) scientists. External support included subject-matter 
experts and scientists from National Aeronautics and Space 
Administration (NASA), the Air Force, The MITRE Corporation, and 
others. In sixteen meetings held over 14 months, the group utilized 
fatigue and sleep scientists, medical experts, and other experts from 
the safety and aviation worlds to help in analyzing the numerous 
fatigue issues and developing viable recommendations.
    The Workgroup decided to develop formal mitigations in order to 
address the hazards and operational risks caused by fatigue. To do 
this, the Workgroup focused on discovering the science and data that 
supported the safety case for each mitigation, with their specific 
focus being the following: to increase the safety of the NAS; to 
improve the health and well being of the workforce; to base any 
findings and recommendations on science and data, and to collaborate 
with internal and external organizations along the way.
    The Workgroup's recommendations were briefed to the FAA 
Administrator in the spring of 2011, shortly before the series of 
incidents. NATCA fully supports their 12 recommendations, and advocates 
adopting all 12 to effectively mitigate the risks associated with 
fatigue.

What Science the Workgroup Relied Upon
    The reality is that ATC operations demand shift work. ATC is a 24-
hour, 7 day-a-week operation (24/7). Fifty-one percent of federally 
operated Terminal facilities are 24/7. One hundred percent of En Route 
facilities are 24/7. Over 3,000 controllers are exposed to midnight 
shifts annually, sitting in dark rooms frequently with little traffic 
to direct. Shift work contributes to cumulative fatigue (overall sleep 
debt), as well as acute fatigue (immediate fatigue that can affect an 
individual at any time of day). Time on task and task intensity also 
contribute to fatigue. Since we cannot eliminate shift work, the 
Workgroup developed formal mitigations in order to address the hazards 
and operational risks caused by fatigue.
    The fundamental question is how does fatigue happen and how does it 
compromise safety? Fatigue refers to a physiological state in which 
there is a decreased capacity to perform cognitive tasks combined with 
an increased variability in performance. There is an established cause 
and effect relationship between the two forces, influenced by multi-
variant fatigue drivers and causes. To correlate the cause and effect, 
the Workgroup developed a multi-layered approach of mitigations that 
fall within six different areas which all interrelate.
    The effects and impacts of fatigue are well documented in many 
industries--from pipelines, trucking, rail, and shipping to the nuclear 
power industry. The physiological and cognitive impacts relate to one's 
ability to stay on task as your accuracy and timing degrade, as you 
experience involuntary micro-sleeps, and as your attention wanes. The 
impacts to individual performance can be numerous, from a loss of 
situational awareness, to an increased risk of operational errors, to 
an overall decline in performance. The cost to productivity can be high 
in terms of both increased absenteeism and higher operational costs. 
Finally, the impact of fatigue on safety is clear: since 1993, over 14 
accidents resulting in 263 fatalities had fatigue as a causal or 
contributing factor.
    Fatigue drivers are clear. There are four of consequence: Circadian 
rhythm; the amount of time since the last sleep period; the quantity 
and quality of one's sleep; and task intensity as a result of workload. 
The primary driver is Circadian rhythm, which is the physiological 
regulator of the human ``sleep and awake'' cycle. Circadian rhythms 
combined with sleep debt, cause sleep pressure, i.e., the urge to 
sleep, especially at night, which creates problems inherent on midnight 
shifts.
    Fatigue causal factors include: (1) workplace elements such as 
schedule, culture, seniority, task complexity, and the physical 
environment; (2) personal elements such as life events and personal 
choices; and (3) individual differences and biological factors such as 
sleep disorders (there are over 70 disorders that influence how we 
sleep) and age. The NTSB uses similar drivers and causal factors as 
their criteria whenever it analyzes for fatigue during a post-accident 
investigation.

Methodology and Findings of the Workgroup
    The Workgroup sought to determine the extent to which ATC schedules 
induce fatigue, and which schedules provide increased cognitive 
performance and opportunity for restorative rest over a six-week 
timeframe. They identified the most widely used schedules and modeled 
110 schedule and sleep permutations to identify risk. They also modeled 
alternative work schedules that increased opportunities for restorative 
nighttime sleep between shifts. Finally, they comparatively analyzed 
modeling results to measure the effect of proposed countermeasures and 
schedule adjustments.
    The Workgroup found that the greatest risk on any schedule is 
during the midnight shift when sleep pressure becomes intense as the 
body is fighting its natural Circadian rhythm. That can be compounded 
by cumulative sleep debt or simply acute fatigue. During that time, 
introducing a sleep opportunity during a shift can mitigate the risk of 
reduced cognitive performance due to fatigue. Proactive sleep prior to 
a midnight shift proved beneficial. Other personal mitigation 
techniques for a recuperative break may include exercise, hydration, 
light exposure, and caffeine.

What the Science-based Workgroup Recommends
    The Workgroup developed 12 recommendations in six topical areas. 
Those areas are: recuperative breaks, scheduling, sleep apnea (SA), 
personal fatigue management, education, and the Fatigue Risk Management 
System. None of these recommendations stands by itself as sufficient to 
adequately mitigate fatigue risks in ATC operations; therefore the 
proposed solution requires the implementation of all of the 
recommendations together, in a comprehensive, layered fashion. The 
Parties have only now begun the process of evaluating and analyzing 
them for their potential impacts on staffing, budget, policy, the CBA, 
and other areas.
    The recommendations and a summary of their related findings are as 
follows:

        1. As fatigue can occur at any time and on any shift, the 
        introduction of a recuperative break during a shift can 
        mitigate the risk of reduced cognitive performance due to 
        fatigue. The Parties recommend that current policy and orders 
        be modified to permit recuperative breaks during relief 
        periods.

        2. Extensive scientific modeling clearly proves that 
        introducing a recuperative break on the midnight shift can 
        mitigate the identified risk of reduced cognitive performance 
        due to fatigue. Re-entry time must be accounted for in all 
        recuperative break planning, execution and management. The 
        Parties recommend the allowance for a recuperative break of up 
        to 2\1/2\ hours.

        3. Quick turns between evening and day shifts reduce 
        opportunities for nighttime restorative sleep. On a 2-2-1, 
        increasing the time between the second evening and the first 
        day shift by one hour increases sleep opportunity and cognitive 
        performance. The Parties recommend the scheduling of a minimum 
        of nine (9) hours between evening and day shifts. This has 
        already been implemented by the Parties.

        4. Scientific modeling shows that increasing night time sleep 
        opportunity during the night prior to the second day shift and 
        subsequent midnight shift results in significant fatigue risk 
        reduction during the midnight shift. However, the placement of 
        the one hour from the reduced shift into a previous evening or 
        day shift has no effect on this risk reduction benefit. 
        Therefore, the Parties recommend that on a 2-2-1 CCW rotation, 
        reduce the day shift preceding the first midnight shift from 8 
        to 7 hours, and begin that shift one hour later, to provide the 
        opportunity for an extra hour of restorative sleep at the end 
        of the night time sleep period.

        5. Per Aerospace Medicine (AAM), 2.2 percent of the ATC 
        workforce has diagnosed sleep apnea, and a minimum of an 
        additional 1.8 percent may be undiagnosed. Perceived non-
        standardized processes, as well as a lack of awareness of sleep 
        disorders and treatments, may result in financial disincentives 
        and unreported sleep apnea in the ATC workforce. The Parties 
        recommend the creation of policies and procedures that 
        encourage self-initiated evaluation, diagnosis and 
        demonstration of initial treatment effectiveness of SA by 
        removal or reduction of economic disincentives.

        6. There is a gap in awareness and understanding of sleep apnea 
        among the controller workforce. Raising awareness and 
        understanding of sleep disorders will reduce the risk to the 
        National Airspace System. The Parties recommend the use of AAM-
        prepared SA education to build Sleep Apnea awareness in ATO 
        workforce, include raising awareness of respiratory coaching to 
        SA patients.

        7. The scope of the sleep apnea issue requires collaboration 
        across respective lines of business. The Parties recommend 
        that:

   AAM to stay current with state-of-the-art in sleep medicine.

   AAM to utilize AASM standards and practices for SA risk 
        factor identification, diagnosis and treatment standards.

   AAM to document the process for medical qualification for 
        individuals at risk for sleep apnea.

   AAM to develop educational materials for the workforce and 
        AMEs.

   AAM to educate AMEs on SA.

        8. Controllers may not fully understand their responsibilities 
        to minimize fatigue, and actions to be taken when they consider 
        themselves too fatigued to safely perform their operational 
        duties. The Parties need to develop policy and education for 
        employees defining responsibilities to minimize fatigue and 
        report fit for duty, and action to be taken when they consider 
        themselves too fatigued to safely perform their duties.

        9. Managers may not fully understand their responsibilities 
        related to interacting with controllers who report that they 
        are too fatigued to safely perform their duties. In order to 
        avoid on-the-job fatigue that threatens safety, the Parties 
        need to develop policy and education for managers that 
        incorporates emphasis on a non-punitive approach when an 
        employee, in accordance with the developed policy, self-
        declares as too fatigued to safely perform operational duties.

        10. Existing controller fatigue awareness training does not 
        comprehensively capture current science, personalize fatigue 
        mitigation strategies, or support practical operational needs. 
        The Parties need to update existing fatigue awareness training 
        to reflect current science and to personalize the application 
        of the training.

        11. A formal Fatigue Risk Management System (FRMS) institutes a 
        continuous, repeatable, collaborative process to identify, 
        analyze and mitigate fatigue risks. The Parties should design 
        and implement a Fatigue Risk Management System (FRMS) within 
        the FAA operational ATC environment.

        12. Retention of organizational knowledge supports a successful 
        transition from the current Fatigue Work Group to the 
        implementation of an approved ATO FRM. The Parties recommend 
        the creation of a transition team composed of current Fatigue 
        Work Group members until the formal FAA FRMS is established for 
        ATC.

    This set of recommendation outcomes flow from the systemic approach 
of a complementing, cross-layered set of prescriptive and non-
prescriptive fatigue risk mitigations. The mitigations would evolve and 
be managed within the formal structure of the FRMS, which 
operationalizes fatigue risk into the FAA decision process and cultural 
fabric. The recommendations equip the Agency to:

        1. Systematically manage ATC fatigue risk;

        2. Reduce acute and chronic sleep debt;

        3. Improve opportunities for nighttime sleep;

        4. Improve ability to obtain restorative sleep;

        5. Allow for the self-declaration of fatigue;

        6. Gather data to support fatigue analysis and mitigations;

        7. Educate the workforce on personal and professional 
        responsibilities in reducing fatigue; and

        8. Support the ongoing adoption of a positive safety culture.

    After the recommendations were presented, the Parties agreed to 
collaboratively examine the implementation considerations for all 
twelve recommendations, with a joint work team that was tasked with 
delivering Questions and Answers within 90 days of their initial 
meeting. Once that group finished their analysis, senior leadership 
from both Parties would determine how to proceed with implementation.

NATCA Recommendations
    NATCA fully supports the implementation of the 12 recommendations 
put forth by the joint Fatigue Workgroup. These recommendations are 
science-based measures to mitigate fatigue and safety risks posed by 
the 24/7 schedule of air traffic control. It is imperative that we act 
quickly to mitigate these risks.
    NATCA recommends that the Agency continue on its current path of 
enhancing the safety culture. Advances in the working environment have 
led to a more open, honest discussion about errors and a more 
transparent process for dealing with those errors in a productive 
manner that deals with root causes rather than punitive responses. As 
one component of that safety culture, NATCA fully supports the 
accurate, precise reporting of all errors and/or safety concerns. While 
the expected rollout of the TARP will add to the increase in reported 
errors, it will help the Agency and controllers perform their jobs with 
more accuracy and a higher degree of safety.
    Thank you again for the opportunity to testify on operational 
errors and fatigue in the workforce. NATCA and the FAA must continue 
working together to mitigate fatigue and safety risks and reduce the 
strain on the National Air Space.

                                Appendix

    An operational error is ``an occurrence attributable to an element 
of the air traffic system in which:

        (1) Less than 90 percent of the applicable separation minima 
        results between two or more airborne aircraft, or less than the 
        applicable separation minima results between an aircraft and 
        terrain or obstacles (e.g., operations below minimum vectoring 
        altitude (MVA); aircraft/equipment/personnel on runways), as 
        required by FAA Order 7110.65 or other national directive; or

        (2) An aircraft lands or departs on a runway closed to aircraft 
        operations after receiving air traffic authorization; or

        (3) An aircraft lands or departs on a runway closed to aircraft 
        operations, at an uncontrolled airport and it was determined 
        that a NOTAM regarding the runway closure was not issued to the 
        pilot as required.

               USA Today OP-ED--Updated 4/17/2011 3 P.M.

            How We're Already Fixing Our Air Traffic System

                   By Randy Babbitt and Paul Rinaldi

    The traveling public rightly expects air traffic controllers to 
make sure their flight safely reaches its destination. We work 
diligently to maintain the trust the American people have in our 
aviation system. But as recently as Saturday, we to uphold that trust.
    After an air traffic supervisor at Reagan National Airport near 
Washington, D.C., fell asleep while two commercial flights landed last 
month, we immediately instituted a review of our air traffic control 
towers. In the last few weeks, we have seen more examples of 
controllers sleeping or being derelict in their duty in Seattle, 
Lubbock, Reno, Knoxville and over the weekend in Miami.
    We cannot and will not tolerate this behavior.
    This week, the FAA is changing long-time controller scheduling 
rules to make sure controllers have more time for rest between shifts. 
We have added staffing at airport control towers and other facilities 
around the country where we had only one controller on the midnight 
shift. Now there are two. We have instituted new hand-off procedures 
for the midnight shift that require contact between radar controllers 
and air traffic control towers to confirm that there is a controller 
prepared to handle each flight.
    These recent incidents have cast doubt on whether our Nation's 
controllers are truly committed to keeping the skies safe. We want to 
tell you they are. We have the safest aviation system in the world, but 
we know we can do better.
    On Monday, we are kicking off our Call to Action on air traffic 
control safety and professionalism. We will be traveling to air traffic 
facilities around the country, to reinforce the need for all air 
traffic personnel to adhere to the highest professional standards.
    Professionalism involves more than just what you do when you're on 
the clock. It means everyone must report to work ready to work. That 
means all air traffic employees must manage their time off 
appropriately and be rested and ready for duty.
    We now understand more about fatigue than we ever did before. The 
FAA has already used the latest fatigue science to propose new rules 
for pilot flight and duty time. Science tells us that working irregular 
day and night shifts without adequate rest periods in between can cause 
chronic fatigue. We are now addressing fatigue in how we schedule our 
controller workforce. The steps the FAA took this weekend are just the 
beginning--we know more needs to be done.
    The FAA will also commission an independent review of our air 
traffic control training curriculum and qualifications to make sure new 
controllers have mastered the right skills and learned the right 
disciplines before they start their careers.
    The National Air Traffic Controllers Association is committed to 
expanding its own Professional Standards program nationwide, which will 
reemphasize for controllers how to maintain the highest degree of 
professional conduct.
    Unfortunately, the events of the last few weeks have tarnished the 
professional and faithful work of thousands of controllers who 
routinely report to their shifts and steadfastly work their stations 
without incident.
    Controllers safely handle an average of 47,000 flights each 
weekday. They direct planes carrying 1.7 million passengers per day. 
And they control air traffic over 15 percent of the world's surface, 
not only over North America but over the Atlantic and Pacific Oceans as 
well.
    We work diligently every day to deliver a flawless performance of 
the air traffic control system. But as is the case with any system 
operated by people, we must have redundancies and back-ups to ensure 
that the system is always safe. And we do.
    As a result, all of the aircraft affected by the recent lapses in 
professionalism remained in contact with air traffic control and landed 
safely. Nonetheless, we are committed to reinforcing our culture of 
accountability in all that we do. There are no simple tasks in 
aviation--every single one is critical.
    We are approaching a complete generational turnover of the 
controller workforce, and in the last 30 years the relationship between 
the FAA and its workforce has been characterized by varying degrees of 
cooperation. But right now our relationship is as strong as it has ever 
been.
    We have an important opportunity to take a step back and look at 
all aspects of our air traffic control system.
    The American public trusts us to perform our jobs and make safety 
the highest priority, each day, year in and year out. We are committed 
to making whatever difficult changes are necessary to preserve that 
trust and to continue to provide the safest and most efficient air 
transportation system in the world.
    Randy Babbitt is Administrator for the Federal Aviation 
Administration. Paul Rinaldi is President of the National Air Traffic 
Controllers Association.

    Senator Cantwell. Thank you, Mr. Rinaldi.
    Dr. Belenky, again, thank you for being here. We look 
forward to your testimony.

              STATEMENT OF GREGORY BELENKY, M.D.,

                RESEARCH PROFESSOR AND DIRECTOR,

             SLEEP AND PERFORMANCE RESEARCH CENTER,

              WASHINGTON STATE UNIVERSITY, SPOKANE

    Dr. Belenky. Thank you. Chairman Cantwell, Ranking Member 
Thune, Chairman of the full committee Rockefeller, 
distinguished members of the Subcommittee, thank you for the 
opportunity to testify on sleep, fatigue and performance in air 
traffic controllers.
    I am Gregory Belenky. I am a physician-by-training and a 
Research Professor and Director of the Sleep and Performance 
Research Center at Washington State University. I joined WSU in 
2004. Prior to that, I served for 29 years on active duty in 
the U.S. Army, developing systems to manage sleep and sustain 
performance in military operations.
    At WSU, we are continuing this work, studying sleep and 
performance in operational environments, operational 
environments in which if the human fails the system fails.
    Chairman Cantwell, it is important for this subcommittee, 
with its critical role in aviation safety, to examine the 
recent incidents in which air traffic controllers have 
inadvertently fallen asleep or deliberately napped while on 
shift.
    Is this a moral failing on the part of a few air traffic 
controllers? Or does it indicate a systemic problem in the 
organizing, staffing and scheduling of air traffic control 
operations?
    I believe it indicates systemic problems, specifically, the 
well-described sleepiness, insomnia and degraded performance 
that is generally characteristic of all nightshift work. Air 
traffic controllers are the same physiologically as any other 
nightshift worker, and the same principles apply.
    What can we learn from these incidents of air traffic 
controllers sleeping on duty? By inadvertently falling asleep 
or deliberately napping on shift, air traffic controllers are 
pointing to a possible problem. They are identifying shifts and 
schedules of shifts that carry relatively higher fatigue risk 
and are in need of fatigue mitigation.
    And, by sleeping on shift, they not only point to the 
problem, they point to a solution as well. The primary 
mitigation for fatigue is sleep, and, in this case, additional 
sleep could most easily come in the form of sanctioned, 
scheduled on-shift napping.
    In the early morning of August 27, 2006, Comair Flight 5191 
crashed on takeoff from Lexington, Kentucky, killing 49 of the 
50 people onboard. The crash occurred at a time when the sole 
air traffic controller on duty was working the last shift of a 
2-2-1 series of shifts consisting of two evening shifts, 2 day 
shifts and finally one nightshift.
    There was an eight- to nine-hour break from the end of the 
second dayshift until the beginning of the final nightshift. 
Unfortunately, this break fell largely in the early to mid 
evening during the so-called ``forbidden'' zone for sleep. So 
the controller was only able to initiate and sustain sleep for 
two to three hours in the late afternoon.
    Comair 5191 crashed at 6:06 a.m., as the captain, first 
officer and the air traffic controller failed to detect the 
plane was on the wrong runway, a runway much too short for a 
successful takeoff.
    A fatigue analysis, including mathematical performance 
prediction modeling, suggests that, at the time of the crash, 
the air traffic controller's performance was impaired by a 
combination of sleep restriction and working at his circadian 
low.
    Given the structure of the 2-2-1 shift series, an on-shift 
nap would have been the only way to increase sleep time in the 
controller during the 24 hours preceding the crash.
    Though the National Transportation Safety Board did not 
implicate fatigue as a cause in the crash of Comair 5191, I 
believe it possible that had the air traffic controller had 
more sleep and been less fatigued he might have detected the 
error in runway choice prior to the attempted takeoff and in 
time to avert the disaster. This is a function of the 
stochastic nature of error, incident and accident, the 
probabilistic element in real-world operations.
    I think that one way to sustain operational performance and 
well-being in air traffic controllers working the nightshift is 
sanctioned, scheduled on-shift napping. We could validate this 
proposed fatigue mitigation/countermeasure by testing the 
effect of sanctioned, scheduled napping on performance and 
vigilance at night-shift operations in select air traffic 
control sites.
    Previous work in air traffic controllers working the 
nightshift has shown that even short, poor-quality naps improve 
alertness and performance.
    As a research scientist, I can describe what the scientific 
evidence suggests is possible and propose ways to develop more 
relevant evidence.
    The members of this subcommittee, as well as labor and air 
traffic control management, must decide what is feasible and 
desirable within the range of possible countermeasures as 
supported by the evidence.
    Thank you, Chairman Cantwell, for the opportunity to 
testify before the Subcommittee. That concludes my remarks. I 
would be happy to answer any questions that you and the members 
of the Committee may have.
    [The prepared statement of Dr. Belenky follows:]

  Prepared Statement of Gregory Belenky, M.D., Research Professor and 
   Director, Sleep and Performance Research Center, Washington State 
                          University, Spokane

    Chairman Cantwell, Ranking Member Thune, and distinguished members 
of the Subcommittee: thank you for the opportunity to testify on sleep, 
fatigue, and performance in air traffic controllers. I am Gregory 
Belenky. I am physician-by-training and Research Professor and Director 
of the Sleep and Performance Research Center at Washington State 
University (WSU), Spokane. I joined WSU in 2004. Prior to that, I 
served for 29 years on active duty in the U.S. Army, developing systems 
to manage sleep and sustain performance in military operations. At WSU, 
we are continuing this work, studying sleep and performance in 
operational environments, environments in which if the human fails the 
system fails. Operational environments include military operations, 
medicine, all modes of air, land, and waterborne transportation, 
security work, first responders, energy generation, resource extraction 
(mining and drilling), financial markets, and industrial production. We 
study normal people under extremes of scheduling. We are supported by 
grants and contracts from the U.S. Department of Transportation, the 
Department of Defense, and the National Institutes of Health, as well 
as state agencies, industry, and philanthropic foundations.
    Chairman Cantwell, it is important for this subcommittee, with its 
critical role in aviation safety to examine the recent incidents in 
which air traffic controllers have inadvertently fallen asleep or 
deliberately napped while on-shift. Questions abound. Is this a moral 
failing on the part of a few air traffic controllers or does it 
indicate a systemic problem in the organizing, staffing, and scheduling 
of air traffic control operations? I believe it is a systemic problem, 
specifically the well-described sleepiness and degraded performance 
that is generally characteristic of all nightshift work--the 
difficulties encountered when trying to work when one should be asleep 
and trying to sleep when one should be awake (Drake and Wright, 2011). 
Air traffic controllers are the same physiologically as any other 
nightshift worker, and the same principles apply. Given the structural 
realities of scheduling, the solution to this problem may lie in 
sanctioned, scheduled on-shift napping when working the nightshift.
    We know that fatigue, operationally defined as degraded 
performance, results from the interaction of sleep loss, circadian 
phase, and workload (McDonald, Patel, and Belenky, 2011; Wesensten et 
al., 2004).
    Performance depends upon total sleep time in 24 hours. Thus sleep 
can be split into two or three sleep periods (a main sleep plus one or 
two naps) and will sustain roughly the same level of performance as a 
single consolidated sleep (Mollicone, et al., 2007, 2008). Simply put, 
naps add to recuperative sleep time. If the main sleep period is 
truncated as it is in shift work, naps can make up the difference.
    The circadian rhythms in task performance and sleep propensity 
parallel the 24-hour circadian rhythm in core body temperature. Task 
performance peaks in mid-evening just subsequent to the peak in the 
circadian core body temperature and troughs in the early morning just 
subsequent to the trough in circadian core body temperature. Twelve 
hours out of phase with performance, sleep propensity troughs in mid-
evening and peaks in the early morning. It is difficult to fall asleep 
and to stay asleep when core body temperature is rising or high and 
easy to fall asleep and to stay asleep when core body temperature is 
falling or low. Hence, the reduced daytime sleep time in people working 
the nightshift and attempting sleep during the day. Sleep is 
particularly difficult in the early to mid-evening, the so-called 
``forbidden zone'' for sleep.
    What can we learn from these incidents of air traffic controllers 
sleeping on duty? By inadvertently falling asleep or deliberately 
napping on-shift, air traffic controllers are pointing to the problem. 
They are identifying shifts and schedules of shifts that carry 
relatively higher fatigue risk and are in need of fatigue mitigation. 
And, by sleeping on shift they are pointing to the solution. The 
primary mitigation for fatigue is sleep. Additional sleep could come in 
the form of sanctioned, scheduled on-shift napping.
    It is a step forward to have two air traffic controllers on duty at 
all times even during slow shifts. However, the full value of this 
increased staffing will likely only be realized if it is leveraged by 
napping on-shift. As commercial airline pilots will tell you, simply 
augmenting flight crews without providing an opportunity for sleep 
isn't much help--it just means three or four tired pilots instead of 
two.
    In the early morning of August 27, 2006, Comair Flight 5191 crashed 
on take-off from Lexington, Kentucky, killing 49 of the 50 people 
onboard. The crash occurred at a time when the sole air traffic 
controller on duty was working the last shift of a 2-2-1 series of 
shifts consisting of two evening shifts, 2 day shifts, and finally one 
nightshift. Working through the night, he was coming to the end of the 
final nightshift of the 2-2-1 schedule when the crash occurred. The day 
shift preceding this trailing nightshift began early the previous 
morning and ended mid-afternoon. The air traffic controller then had 
the regulation-mandated 8-9 hours off duty before going back on duty in 
the late evening for the nightshift. He managed only 2-3 hours of ``not 
real good'' sleep in the late afternoon. He then remained awake through 
the evening. His sleep was truncated because the bulk of his sleep 
opportunity fell in the early to mid-evening, the so-called ``forbidden 
zone'' for sleep. With respect to the relationship between sleep and 
circadian physiology, the controller took the maximum possible 
advantage of the sleep opportunity he was given. He went back on duty 
at 11:30 p.m. with his shift projected to end at 7:30 a.m. Comair 5191 
crashed at 6:06 a.m. as the captain, first officer, and the air traffic 
controller failed to detect that the plane was on the wrong runway, a 
runway much too short for successful take-off. A fatigue analysis, 
including mathematical performance prediction modeling, suggests that 
at the time of the crash the air traffic controller's performance was 
impaired by a combination of sleep restriction and working at his 
circadian low (see Figure 1) (Pruchnicki, Wu, and Belenky, 2011). 
Having another controller on duty to enable alternating on-shift naps 
would have been the only way to increase sleep time in the controller 
on the 2-2-1 schedule during the 24 hours preceding the crash. Though 
the National Transportation Safety Board did not implicate fatigue as a 
cause, I believe that had the air traffic controller had more sleep and 
been less fatigued he might have detected the error in runway choice 
prior to the attempted takeoff and in time to avert the disaster.



    Figure 1: Performance prediction for the air traffic controller on 
duty during the Comair 5191 crash. Note that his predicted 
effectiveness at the time of the crash, marked by the asterisk, was 71 
percent (from Pruchnicki, Wu, and Belenky, 2011).

    Twenty years ago, then NASA scientists Curt Graeber and Mark 
Rosekind conducted a pioneering study that demonstrated the 
effectiveness of scheduled cockpit napping in sustaining performance 
and vigilance in flights across the Pacific (Rosekind, et al., 1994). 
In this study, on-shift napping improved performance.
    Recently, Charles Czeisler and colleagues in the Harvard Work Hours 
Health and Safety Group carried out a remarkable study of rates of 
medical errors associated with extended work hours and sleep loss 
(Landrigan, et al., 2004; Lockley, et al., 2004). They found that when 
publically-funded physicians in post-graduate residency training were 
decreased from an 85-hour to a 65-hour work week, and, as a result, 
obtained more sleep, they experienced a one-third reduction in the rate 
of serious medical errors that included a five-fold decrease in the 
rate of serious diagnostic errors. In this study, limits on work hours 
increased sleep and improved performance.
    American, Continental, and Delta Airlines are currently conducting 
studies in pilots flying augmented (4-pilot) long-range flights. From 
these and other studies, it is apparent that pilots are able to take 
advantage of the on-board crew bunk facilities during cruise for rest 
and sleep. And, they do sleep. This sleep is on-shift napping, 
sanctioned by the FAA and paid for by the airlines.
    I expect that an effective way to sustain operational performance 
and well-being in air traffic controllers working the nightshift is 
sanctioned, scheduled on-shift napping. We could validate this proposed 
countermeasure by testing the effect of sanctioned, scheduled napping 
on performance and vigilance in nightshift operations in select air 
traffic control sites. Previous work in air traffic controllers working 
the nightshift has shown that even short, poor quality naps improve 
alertness and performance (Signal et al., 2009).
    As a research scientist, I can describe what the scientific 
evidence suggests is possible and propose ways to develop more relevant 
evidence. The members of this subcommittee, as well as labor and air 
traffic control management, must decide what is feasible and desirable 
within the range possible countermeasures as supported by the evidence.
    Thank you, Chairman Cantwell for the opportunity to testify before 
the Subcommittee. I would be happy to answer any questions that you and 
the members of the Committee may have.

References
    Drake, C.L. and Wright, K.P. (2011) Shift work, shift-work 
disorder, and jet lag. In Kryger, M., Roth, T. and Dement, W.C. (Eds.) 
Principles and Practice of Sleep Medicine, 5th Edition, pp. 784-797.
    Mollicone, D.J., Van Dongen, HPA, and Dinges, D.F.. (2007) 
Optimizing sleep wake schedules in space: Sleep during chronic 
nocturnal sleep restriction with and without diurnal naps. Acta 
Astronautica, 60:354-61.
    Mollicone, D.J., Van Dongen, HPA, Rogers, N.L., et al., (2008) 
Response surface mapping of neurobehavioral performance: Testing the 
feasibility of split sleep schedules in space operations. Acta 
Astronautica, 63:833-40.
    Landrigan, C.P., Rothschild, J.M., Cronin, J.W., Kaushal, R, 
Burdick, E, Katz, J.T., et al., (2004) Effect of reducing interns' work 
hours on serious medical errors in intensive care units. N Engl J Med. 
351: 1838-48.
    Lockley ,S.W., Cronin, J.W., Evans, E.E. et al., (2004) Effect of 
reducing interns' work hours on sleep and attentional failures. New 
Engl J Med. 351: 1829-36.
    McDonald, J., Patel, D., and Belenky, G. (2011) Monitoring Sleep 
and Performance in the Workplace. In Kryger, M., Roth, T. and Dement, 
W.C. (Eds.) Principles and Practice of Sleep Medicine, 5th Edition, pp. 
775-783.
    Pruchnicki, S, Wu, L, and Belenky, G. (2011) An exploration of the 
utility of mathematical modeling predicting fatigue from sleep/wake 
history and circadian phase applied in accident analysis and 
prevention: The crash of Comair Flight 5191. Accident Analysis and 
Prevention, 43(3): 1056-1061.
    Rosekind, M.R., Graeber, R.C., Dinges, D.F., et al., (1994) Crew 
factors in flight operations IX: Effects of planned cockpit napping on 
crew performance and alertness in long-haul operations. NASA Technical 
Report 108839; http://human-factors.arc.nasa.gov/zteam/PDF_pubs/
Flight_Ops/Cockpit_Rest_IX/c-rest_opsumm.pdf.
    Signal, T.L., Gander, P.H., Anderson, H., and Brash, S. (2009) 
Scheduled napping as a countermeasure to sleepiness in air traffic 
controllers. J. Sleep Res., 18: 11-19.
    Wesensten, N.J., Belenky, G., Thorne, D.R., Kautz, M.A., and 
Balkin, T.J. (2004) Modafinil versus caffeine: Effects on fatigue 
during sleep deprivation. Aviat Space Environ Med, 75, 520-525.

    Senator Cantwell. Thank you, gentlemen. Thank you, Dr. 
Belenky, and thank you all for your testimony.
    Dr. Belenky, I think I'll start with you on this last point 
that you just made about what is the optimum schedule you're 
talking about within the framework of what exists today, but is 
there an optimal schedule to minimizing fatigue?
    Dr. Belenky. Well, Senator Cantwell, yes. The optimal 
schedule is daytime work and 8 hours of nighttime sleep. 
Unfortunately, there is no good solution for nightshift work. 
Many things have been tried--stimulants, bright light, 
melatonin, various behavioral----
    Senator Cantwell. I'm referring to the fact of these 2-2-1 
schedules of----
    Dr. Belenky. Ah, the 2-2-1.
    Senator Cantwell.--of day and nightshift. Do organizations 
that have a strict nightshift workforce have a better way of 
dealing with this issue as opposed to this mix of day and 
nightshift?
    Dr. Belenky. The rapid turn on the 2-2-1 is particularly 
troublesome, but there are problems with full-time nightshifts, 
rotating nightshifts, forwardly rotating, backwardly rotating. 
None of these are good. Early starts pay a huge penalty on 
sleep time and performance degradation. They're almost as bad 
as working permanent nights.
    So, there is no optimal solution. There are many partial 
fixes that improve performance to a degree, but no one-size-
fits-all schedule that will work under all circumstances. I 
hope that's--Is that----
    Senator Cantwell. Well, yes, thank you. I wanted to clarify 
that, and then your testimony obviously talks about what you 
think some of the remedies are within that framework.
    I want to go back to--Mr. Scovel, in your testimony, I 
wanted to--Do you have a sense why there has been this 
significant increase in operational errors since fiscal 2010? I 
know Mr. Rinaldi doesn't want us to look at 2010, but did you 
have a sense of this?
    Mr. Scovel. Thank you, Madam Chairman. Yes, certainly the 
numbers reported by FAA do show an increase in operational 
errors from 2009 to 2010, a 53-percent increase. In fact, the 
1,234 errors in 2009 rose to 1,887 in 2010. The question is 
why?
    As our written statement shows, we don't know. Neither does 
FAA at this point either. It could be better reporting 
practices, and we think that probably accounts for some of it. 
It could be an increase in the number of operational errors 
itself, and then, through better reporting practices, that 
increase is also captured.
    The reporting practices that I'm referring to are what both 
Mr. Rinaldi and Administrator Babbitt spoke of earlier, and 
that is the Air Traffic Safety Action Program. They believe it 
has encouraged an atmosphere of self reporting minus possible 
professional repercussions for controllers submitting reports.
    However, Mr. Babbitt has stated that reports of operational 
errors submitted through ATSAP are not included in those 
counts. So that cannot explain the increase.
    The Traffic Analysis and Review Program that you, Madam 
Chairman, referred to earlier in your statement may explain 
part of it. And, in fact, as we sliced and diced some of the 
numbers, we found an 86-percent increase in reported 
operational errors at TRACON facilities from 2009 to 2010. TARP 
may account for part of it.
    However, we're puzzled by the fact as well that en route 
air traffic control facilities reported operational errors 
increased 39 percent from 2009 to 2010, and at en route 
facilities, where they have had a program like TARP, an 
automated detection and reporting tool, in place for some time. 
So that would indicate that, at least at en route centers, 
there's an absolute and bottom line increase in operational 
errors. We don't know why.
    In our visits to air traffic control facilities as part of 
our audit work, we have discussed this with managers and on-
line controllers and they've told us some of them believe that 
it's due to the increased number of controllers in training, 
and that stands, perhaps, to logic.
    Others have told us that they think it's attributable to 
controllers at the midpoint of their career who are beyond the 
training stage when they might be more careful in each and 
every action and who have become somehow more complacent. We 
just don't know.
    But, Madam Chairman, you have kindly asked our office, and 
we have requests as well from the House, to review both the Air 
Traffic Safety Action Program as well as FAA's LoSS Index, 
which will attempt to capture all such losses of separation, 
categorize them, and, we hope, attempt to gather some data on 
those, so that it can be properly analyzed and corrective 
measures prescribed.
    Senator Cantwell. And, Mr. Scovel, just following up on 
that last point before I turn it over to my colleagues, do you 
have any information in data as it relates to that separation--
loss-of-separation issue as it relates to this fatigue issue, 
any information about that today?
    Mr. Scovel. I'm sorry, Madam Chairman, are you referring 
specifically to the LoSS Index or----
    Senator Cantwell. Loss of separation of flights and this 
issue of fatigue. Have you found any issues of how those are 
connected at this point in time?
    Mr. Scovel. In 2009, at the request of Senator Durbin, we 
examined potential fatigue factors at the three main air 
traffic control facilities in Chicago. We identified scheduled 
overtime, minimal time for rest between shifts with a counter-
rotational shift pattern with progressively earlier times, as 
well as high demands for on-the-job training on the part of 
veteran controllers at those facilities as key factors. 
Controllers reported a degradation in their performance and 
increased fatigue as a result of that. We did not link those 
specifically to operational errors.
    Senator Cantwell. So we don't have a link between these two 
issues at this point.
    Mr. Scovel. We don't----
    Senator Cantwell. As it relates to operational errors and 
air separation. OK.
    Mr. Scovel. Yes.
    Senator Cantwell. Senator Rockefeller.

           STATEMENT OF HON. JOHN D. ROCKEFELLER IV, 
                U.S. SENATOR FROM WEST VIRGINIA

    The Chairman. Thank you, Madam Chair.
    Mr. Scovel, I think you said, the first time around, that 
it's very important to place people in the right airport, right 
air tower, and that large and complex ones for those who are 
just in training or relatively new into it is maybe not a good 
idea.
    And so that makes me want to ask Randy Babbitt the question 
of how he handles that. How are people assigned?
    And you may want to comment on this, Mr. Rinaldi. How are 
people assigned?
    And it strikes me as a very smart point that he made. It 
doesn't cure a lot of problems, but it sort of creates a 
baseline of at least an attempt at prioritizing.
    Mr. Babbitt. To answer the question, the placement has been 
made far more rationally today. Under the new agreement that we 
have with the air traffic controllers, we now can provide 
incentives to air traffic controllers to move to the more 
complex facilities.
    We didn't have that opportunity under the last agreement, 
and, therefore, we often had a situation where a vacancy would 
come open in the most complex of facilities and no one would 
bid it. So we were forced to assign people fresh out of 
training, not necessarily in accord with our wishes, but simply 
because it was the only way to fill a vacancy.
    And that was unfair to the controllers involved. It was 
unfair to the controllers doing the training. It was unfair to 
the facility. That has been remedied, and, today, controllers 
can and will bid the more complex facilities.
    I think in any business venture, when you assign people, 
you certainly pay attention to how they do in their training. 
We like to think that everyone who graduates from the academy 
is suitably qualified to operate anywhere in our system.
    We also try to honor their wishes. If someone grew up in 
Seattle and wanted to be an air traffic controller in Seattle 
and that is an available vacancy, we would let them bid into 
Seattle or New York or wherever their home might be. People 
often bid to work where they grew up, and so, to the extent we 
can, we honor these preferences as well.
    I also want to make one comment. There seems to be some 
thought that the rate of people who are in training is rising, 
and it's not. It's falling. With normal turnover we have 
historically had about a 25-percent rate of the controllers 
throughout the FAA in training. That includes people who 
transition.
    If I were a controller yesterday in the Cleveland center 
and I moved to the Denver center, when I move there, I am 
considered an in-training controller. It takes me a while to 
learn that system. I may have been with the FAA for 20 years. I 
am a very well-qualified controller, but, in that transition, I 
am treated as a controller-in-training, and, therefore, I count 
in these ranks.
    A year ago, the number was 30 percent. Today, it's 25 
percent. So we are reducing the number of retirements. There 
was a surge. We had an exceptional number of retirements for 3 
or 4 years after the last agreement. The previous agreement led 
a lot of people to retire.
    We had an enormous surge in retirements, and, therefore, 
were dealing with a corresponding increase in the training. And 
I believe the Inspector General noted that we had a 25-percent 
increase in cost. We had about a 35-percent increase in 
training which would account for that increase in cost.
    The Chairman. My time is about to run out. I'll get to you 
next time, Mr. Rinaldi.
    As a background for all of this, the House has passed a 
budget which would take everybody--FAA--back to 2008 levels.
    Now, I'm just trying to consider what the effect would be 
on what we've discussed so far, what the flexibility would not 
be for you and others because of the desperate--you know, the 
fact that a lot of these people would be laid off and you'd be 
dealing in a winter wonderland.
    I'd like you to explain how, if we went back to 2008 
levels--and, believe me, I'm not going to let us do that, but 
if we do, the world needs to know what would happen on your 
watch about this.
    Mr. Babbitt. Well, let me start with saying I'm not going 
to budget safety. Safety will be maintained at the level that 
it is today. We're going to inspect all the airplanes. The 
facilities will operate. We don't have the option of shutting 
down radar for 10 percent of the time because we had a 10-
percent budget cut. So we will maintain the level of safety.
    But an area where we think we will feel impact, for 
example, is in flight standards, where we have over 850 
requests awaiting approval today. These are airlines that are 
buying new equipment. They're pioneering new routes. They're 
doing a lot of new things, maybe opening new stations. These 
all require our certification.
    We have 2,400 items--safety items that are in the queue to 
be certified. This includes new wingtips, new electronic 
equipment, advanced engines, all needing certification. These 
are all objects that would make the aviation system better, 
cheaper, more efficient, more environmentally friendly.
    The Chairman. But nobody can do anything until they get 
certified.
    Mr. Babbitt. They can't do anything, nor will they employ 
people who would build them, which leads to the final point, 
NextGen itself. NextGen becomes very seriously threatened.
    I recently read a private sector report that said that if 
we delay NextGen for 5 years, it will cost $148 billion worth 
of the potential value that we get by building the system now. 
So to delay it 5 years has an enormous impact.
    The Chairman. I thank the Chair.
    Senator Cantwell. Thank you.
    Senator Thune.

                 STATEMENT OF HON. JOHN THUNE, 
                 U.S. SENATOR FROM SOUTH DAKOTA

    Senator Thune. Thank you, Madam Chairman. And I want to 
thank you for holding this important hearing today and for our 
witnesses for testifying.
    And this is my first hearing as Ranking Member, so I'm 
looking forward to tackling the important issues that fall 
under this subcommittee's jurisdiction.
    I think, in most circumstances, our constituents always 
want us to be able to share their experiences, and when it 
comes to flying, most of us do. Most of us are frequent flyers, 
so we can certainly identify with the challenges that people 
face in traveling. And I want to work with my colleagues on 
this committee to ensure that our consumers in this country 
have access to affordable, safe and timely air service.
    Our Nation's airspace and the almost 25-million square 
miles of oceanic air space that the FAA is tasked with 
monitoring involves roughly 15,000 air traffic controllers and 
almost 1,300 civilian contract controllers and more than 9,500 
military controllers. So, together, they ensure that our Nation 
has one of the safest aviation systems in the world.
    But, as we have found from recent reports, there's still a 
lot of work that needs to be done, and so I appreciate hearing 
some of the steps that are being mentioned today.
    And I wanted to take up a question, if I might, Mr. 
Babbitt, with you regarding NextGen. You mentioned it, and I'm 
aware that FAA is in the early stages of implementing some 
major advances in air traffic control management with the next-
generation air traffic control system that the system will use 
technological advances to make aviation safer, more efficient.
    In some cases, aircraft will also be flying closer together 
more safely, that's the plan. To what extent, if at all, do you 
see the NextGen system preventing or reducing incidents that 
could be caused by controller errors?
    Mr. Babbitt. Well, one of the advantages that the NextGen 
system brings us is very enhanced and increased situational 
awareness, so that the display in the cockpit will show all the 
aircraft around your aircraft. So you'll have essentially the 
same display that the air traffic controller has, and it's just 
simply a backup system.
    As hard as we try, there have been situations where radar 
fails. Sometimes it's for a few moments, a few seconds, but 
when airplanes are closing at a rate of eight miles a minute 
each, a combined closure rate of 16 miles a minute, 20 seconds 
is a long time. Having situational awareness improved through 
warnings that pilots would get from that type of better 
situational awareness and warning technology could be a huge 
lifesaver.
    So, yes, it gives us a much better and timely situational 
update. People don't think about it, but where there are long-
range radar sweeps--Paul could tell you better than I, I think 
it's between 9 and 12 seconds between updates. Airplanes go a 
long ways in 12 seconds. So that's when they update themselves 
again. Updated situational awareness is instantaneous. They're 
constantly showing the accurate position of where the aircraft 
are.
    Senator Thune. Is there anything in the NextGen system that 
could be improved that might help alleviate problems that are 
caused by controllers falling asleep?
    Mr. Babbitt. Well, the issue there, I think, is being 
managed a little differently. We're taking a real hard look and 
working with our colleagues at NATCA to work through the 
scheduling process as fatigue mitigation and so forth.
    The technology of NextGen is going to be more effective in 
terms of providing everybody with better situational awareness 
of where the traffic is. But, no, I don't see the relationship 
to alertness.
    Senator Thune. If anybody else on the panel would want to 
respond to that, feel free to. I was directing that to Mr. 
Babbitt.
    But I also wanted to point out, I guess, over the next 10 
years we're going to be looking at 11,000 new controllers being 
hired and trained. And so I would direct this to you, Mr. 
Babbitt as well, but are there any programs in place that would 
be able to identify who might be more adept or who might have 
the greatest difficulties at working midnight shifts? Is there 
any way that you can identify those types of things when you're 
evaluating personnel?
    Mr. Babbitt. Dr. Belenky could probably shed a little more 
light on profiling, but, no. Just in terms of overall training, 
one of the things that we did this morning, is that I kicked 
off a sort of blue ribbon panel, if you would--a group that 
we've selected in this overall review of air traffic control 
training.
    And this panel of five is going to look at every segment of 
how we hire, how we train, how we requalify our controllers. 
Are we teaching them the right things? Is the curriculum right? 
Are we getting the right ratios through our school? And these 
are all individual experts in their field. So I'm looking 
forward to their report to us to help us improve the training 
of the controllers.
    Senator Thune. Mr. Belenky.
    Dr. Belenky. Senator Thune, people who are morning types do 
not do as well in nightshift work as people who are evening 
types. There is actually a difference in their circadian rhythm 
phase, with the trough being earlier in the morning for morning 
types, later in the evening for evening types. This is a 
physiological trait difference between people.
    Evening types do better at nightshift work, Also, younger 
people do better. As people get older, sometimes someone who 
tolerated shift work very well ceases to tolerate it as well or 
to tolerate it at all. This may be because, as we get older, we 
shift toward becoming more and more a morning type. Therefore, 
this is an issue, and there are physiological differences that 
do speak to people's ability to do this.
    Senator Thune. Well, it just seems like that with that kind 
of information that managers might be in a better position to 
schedule and mitigate potential issues for controllers before 
they happen, if that kind of information and data is available.
    Dr. Belenky. Yes.
    Senator Thune. And it sounds like it is.
    Dr. Belenky. It is. Yes, this is accepted within the field.
    Senator Thune. Madam Chair, my time has expired, so I'll--
--
    Senator Cantwell. Thank you. Senator Lautenberg.

            STATEMENT OF HON. FRANK R. LAUTENBERG, 
                  U.S. SENATOR FROM NEW JERSEY

    Senator Lautenberg. Thank you, Madam Chairman. I was 
beginning to feel kind of lonesome here, because we introduced 
the Chairperson, then the Chairman of the Committee, the 
Ranking Member and then members of the Committee. So I am 
multiples here, members of the Committee.
    And I'm glad that we have a chance to have this exchange, 
really important, and when we look at the numbers that fly 
every day and how good the performance is of the controller 
force, it's really remarkable.
    But the very obvious glitches that are in here, when you 
look at, now, six incidents in which air traffic controllers 
and supervisors were caught sleeping on the job, forcing pilots 
to land planes with no assistance--and I understand, Mr. 
Babbitt, that you're taking steps to ensure that there are at 
least two people, if I'm correct, in a tower at all times--but 
I wonder, in the processing of appointments to various 
stations, whether--are there any prohibitions against second 
jobs?
    I'm sure a lot of people enjoy second-job income, among the 
controllers as well as other people in the workforce. Are there 
any rules that say, Look, you can't have strenuous exercise 
before you come to work? And that has to be a pledge. I don't 
know how you monitor it. But the fact of the matter is if 
someone just a five- or a ten-mile run and then comes to work, 
could be headed for a very serious problem. Dr. Belenky.
    Dr. Belenky. Thank you, Senator Lautenberg.
    The main determinant of sleep time is work hours. So, if 
you add to the normal work hours with a second job with other 
employment, you cut into your sleep time. In the factors 
determining sleep time, first is work hours, second is travel 
time, including dropping people off and picking people up and 
commuting, and third is family and community responsibilities.
    Senator Lautenberg. Right. But with all of those things, I 
mean, to answer the question as I put it, I mean, how do you 
regulate a behavior? Because that, obviously, has to do with 
sleep.
    Now, I know when I get older, I'll probably--as you said--
will need more sleep. Right now, I'm good, but, anyway----
    Senator Cantwell. I think he said you would just be getting 
up earlier.
    Senator Lautenberg. Yes. We can't continue this dialogue, 
but----
    And I am so proud of our workforce, Mr. Babbitt, but, as 
you know, and there are thousands and thousands of really 
well--good movements and no problems and so forth.
    But it's not the good things you do. That's expected. It's 
the bad things that happen that were focused on, because one 
incident can be one far too many, and we have to be careful 
with that.
    So, you know, we had a major assault on controllers some 
years ago where the whole force was terminated virtually and 
had to rebuild.
    Now--And I hear you ask a plea for sufficient budget to 
take care of your responsibilities. Now--But then, on the other 
hand--and I like what you said, that safety is the most 
important issue and there will be no compromise on safety, but 
how do these things come together?
    If you don't have enough money in the budget, it's pretty 
hard to say, Well, OK, we're all going to do safety measures, 
and the greatest safety would be to spread the hours out, 10 
hours between jobs or whatever that is. There's an 
inconsistency there.
    And I think that, not unlike the military, I mean, when we 
send people to the front, we have to have enough bullets for 
them to carry. And if we send people up in those towers, those 
jobs are equally important, because a mistake could be 
unacceptable under any condition. So how does that work out, 
the budget and safety?
    Mr. Babbitt. Senator, I'll expand a little bit on the 
comment that I made. I mean, you're asking me to make somewhat 
of a Sophie's choice, and I indicated that we would not 
compromise safety, and we won't.
    We have a very dedicated workforce, including the air 
traffic controllers. We're going to adequately staff and man 
our facilities and make certain that they have the rest they 
need, the education they need and so forth.
    But what I did indicate was there are areas that are more 
discretionary, for example, in the certification area. We're 
looking at that right now. There are three different facilities 
being proposed to be built on the East Coast of the United 
States. One is Boeing. One is HondaJet, and one is Embraer. All 
three of these facilities propose to hire anywhere from 1,500 
to 4,000 employees, and each of those facilities has to be 
certified by the FAA.
    Now, is safety going to be compromised if I build one 4 
months later? No, but I would suggest to you I think the 
American public would be far better served by building that 
plant and putting 4,000 people to work 4 months earlier, rather 
than for me to be lacking the 10 people necessary to inspect 
the plant.
    So we're being forced to make some decisions, some 
discretionary spending decisions that I think--you know, 
there's a fairly significant business case that would support 
the request that we've made.
    And I appreciate that all of us want to do better. We want 
to do more with the funds that we have. I think we're very good 
shepherds of the taxpayer dollar. I can point you to savings we 
have achieved. We have undertaken oversight programs within the 
FAA. We've saved $560 million in the last 5 years, and that's 
money that we invested in programs and didn't ask anyone for 
more money. We funded those from internal savings.
    We're going to save $85 million this year, much of it from 
IT consolidation. We've got plans going forward to share our 
services better to be more efficient.
    But, at the end of the day, not having the funds that we're 
looking for will have consequences. It won't be safety, but it 
will have consequences. NextGen will be delayed.
    Senator Lautenberg. Well, it's going to be someplace. I 
mean, you can't get more liquid in a quart bottle than the 
quart was intended to hold.
    And I don't know how, Madam Chairman, that we can say, OK. 
Build additional airplane-building facilities, bring more 
airplanes into the system, and not be guaranteed that we have 
enough funds to supply the appropriate number of controllers.
    And there ought to be a formula established that says, OK. 
You want to cut the funds that go into the FAA, OK, then here's 
how many controllers we have. And we say there can only be X 
number of airplanes in the sky, so that there isn't a question 
about--this tug of war that you find yourself in and that we 
find ourselves in, where, oh, it's going to be cut, cut, cut. 
When you cut too much, the blood starts running, and that's 
what we have to be careful for.
    Thank you very much. And thank you all very much for your 
testimony.
    Senator Cantwell. Thank you, Senator Lautenberg.
    Mr. Babbitt, I'd like to go back to the questioning that I 
was asking Mr. Scovel about operational errors and just trying 
to understand whether you have any purview on this as it 
relates to this year. Are we seeing the same trajectory? Do we 
have any information? Is this year better than last?
    Mr. Babbitt. Let's start with the fact that--I'm as 
concerned with an increase of operational errors as anyone. 
That's not a good thing, but on the other side of it, I am 
pleased. We tried to change the culture. We want people to 
report everything.
    We are now focused on some of the culture changes. We've 
asked our supervisors to be more proactive. And while ATSAP 
reports in and of themselves are excluded from the reporting, 
often operational errors are dually reported.
    I'm your supervisor. You make an error. I see it. You file 
ATSAP. I still file it as an operational error. So there's no 
prohibition on both of us noting that operational error.
    Senator Cantwell. But you're not saying that's double 
accounting there.
    Mr. Babbitt. No, no. No, but the comment was made that the 
ATSAP reports aren't counted, and I'm suggesting to you, that 
sometimes they are, in another fashion. Someone else is going 
to file the report about the incident.
    The other thing that I think we should pay attention to is, 
yes, there has been a dramatic increase, but the A errors--just 
use the A's--the most serious errors; these went from 37 in 
2009 to 43. That was out of 133 million operations. And we saw 
an increase of seven. I don't like an increase of seven, and I 
want to know why those happened, but that's a very small 
percentage of error increase.
    The lion's share of the increase of errors comes down in 
the less significant categories, the C's and the D's. These are 
operational errors. This means that someone who we wanted to 
have five-mile in-trail spacing had 4.9 spacing for a minute. 
That's all. We put that margin there for a good reason. We 
don't want people getting inside of 5 miles.
    But there was nothing at risk here. They simply violated 
the parameters we put around, and I want to know why. So we 
take this increased data and work it in and revise our 
training.
    And, by the way, as we go forward, we're going to get more 
increases in error reporting as we capture more and more 
electronically. I think Mr. Rinaldi will tell you, when you 
look at a radarscope that's scanning 50 miles, can you look at 
it and tell that's 5 or 4.9? No.
    Electronically, you can, and as this TARP-type reporting 
comes in, we're going to see an increase in error reporting, 
and that's a good thing. I want to know why those errors are 
occurring, so we can address them.
    Senator Cantwell. Can you talk about the A group, which is 
the most severe classification of error and what the 
methodology is? Is this subjective? Is this an objective 
process and----
    Mr. Babbitt. No, these errors are ranked A, B, C and D. 
Loss of separation is essentially what we're talking about. 
We----
    Senator Cantwell. And A is----
    Mr. Babbitt. The most significant loss of separation.
    Senator Cantwell. Which is?
    Mr. Babbitt. Well, depending on whether it's an en route 
environment, whether it's in the TRACON final approach, you 
know, each of these are different. For example, over the ocean, 
we separate airplanes with 50-mile, in-trail separation. 
Because we can't see them, we require them to report where they 
are.
    Across the United States, they can go to 20-mile in-trail 
separation. In a TRACON, they can be down to five, and, on 
final approach, because the radar is better, or, when they can 
see each other, we can tolerate three miles. So----
    Senator Cantwell. And on this increase of seven over the 
previous year, do you know which of those they were, whether it 
was----
    Mr. Babbitt. I can get back to you. I don't have that in 
the top of my head, but I certainly could get back to you with 
that data in each particular event.
    [The information referred to follows:]

    In Fiscal Year 2010 there was an increase of six (6) category A 
operational errors from Fiscal Year 2009 (FY 2009--37 and FY 2010--43). 
The increase for Fiscal Year 2010 was within the terminal environment 
with a separation requirement of 3 nautical miles lateral or 1,000 feet 
vertical.

    But my point is that thousands of the increases were down 
in the C's, not significant. They were operational errors. They 
were a loss of separation, but not the significant losses. The 
significant losses are the ones that would really concern us. 
They all concern me. I just wanted to make that distinction.
    Senator Cantwell. And do you know if any of them were 
proportionally more operational errors during the midnight 
shift than other shifts?
    Mr. Babbitt. I don't know the answer to that, but, again, I 
can get back to you. We certainly can get the time and location 
of each event.
    [The information referred to follows:]

    No, there are not proportionately more operational errors during 
the midnight shift as compared to other air traffic controller shifts. 
For Fiscal Year 2010 there were 68 operational errors between the hours 
of 9:30 p.m. and 5:30 a.m. as compared to 1,819 for all other times.




    Senator Cantwell. Because I think that's one of the 
questions that we're trying to ascertain here. We're seeing 
this increase of operational errors. It is very concerning.
    Mr. Babbitt. Sure.
    Senator Cantwell. And we obviously have this issue of 
fatigue in the workplace and----
    Mr. Babbitt. Yes, I would----
    Senator Cantwell. And they're both very concerning.
    Mr. Babbitt. Sure.
    Senator Cantwell. But being related to each other would 
make us even more concerned.
    Mr. Babbitt. Yes, I would----
    Senator Cantwell. And so----
    Mr. Babbitt. Common sense would direct me to suggest that 
probably not, because the traffic drops off significantly in 
the evenings. These operational errors tend to happen in high-
volume situations.
    Senator Cantwell. So is the air traffic controllers' 
schedule and fatigue considered causal factors for operational 
errors or is that--Do you know, Mr. Scovel?
    Mr. Scovel. I'm sorry, you're asking, Madam Chairman----
    Senator Cantwell. The FAA lists causal--do they list causal 
factors for each operational error?
    Mr. Scovel. When FAA launches its investigatory process, 
subsequent to each operational error, there are a series of 
questions that are asked. We believe that those questions need 
to be better refined and the data needs to be much more 
precise.
    For instance, I referred earlier to our review of potential 
fatigue factors at the Chicago area air traffic control 
facilities, and in reviewing operational error reports at that 
location, we were looking specifically for the degree to which 
fatigue was accounted for in the investigation. And we found, 
in too many instances, a cursory description of what the 
controller had experienced that might lead an observer to think 
he might be fatigued.
    For instance, the report form will ask what shift. That's 
entirely relevant. The controller, in some instances, reported 
simply ``rotation.'' The reports did not always indicate which 
shift or which day in the 2-2-1 rotation.
    With better attention from management, and a better list of 
questions to begin with, better data can be obtained. Better 
data, with proper analysis, will yield better corrective 
actions and reduce the risk to the flying public.
    Senator Cantwell. Thank you. Thank you.
    Chairman Rockefeller.
    The Chairman. Thank you, Madam Chair.
    Mr. Rinaldi, I think it's only fair that you get to talk. 
So let me ask you a couple of questions.
    Number one, this has not been answered, and I'm ashamed to 
say I don't know the answer myself. Please tell me that an air 
traffic control person cannot hold two jobs during the course 
of the day.
    Mr. Rinaldi. It's not prohibited, and under the imposed 
work rules of 2006 and payrolls of 2006, many of the new air 
traffic controllers were holding down two and three jobs to 
make ends meet because of the cut in pay.
    I applaud the Administrator for really getting--and the 
Secretary of Transportation--for putting a lot of focus on 
getting us back to a fair collective bargaining agreement.
    And I'm not sure what the number is, if anyone is holding 
two jobs down at this point.
    The Chairman. You said two and sometimes three?
    Mr. Rinaldi. Sometimes three they were, from 2006 to 2009, 
to make ends meet.
    The Chairman. That's stunning to me, because I think that 
Dr. Belenky can do all of the magic he wants, but he can't 
overcome that one. And I'll come to you in a minute, sir. But 
that's an enormous statement. It's an enormous statement. To 
me, that's like asking for trouble.
    Mr. Rinaldi. Fatigue is----
    The Chairman. Your response would be, Well, they don't have 
any choice. They've got to make a living and they've got mouths 
to feed.
    That's where we, again, get into the question of the budget 
not affecting this. Safety comes first. Well, the budget's 
going to affect this. It's going to affect pay increases or 
non-pay increases. Just like not having NextGen makes people's 
life much more complicated.
    On the other hand, it makes it much better, because they 
can see farther out and have much more accurate spacing 
readings.
    But how can this happen? Has this just always been the 
case? How do you make the case that this doesn't cause 
sleeplessness or bad judgment?
    Mr. Rinaldi. Well, actually, from the years of 2006 to 
2007, we were talking exactly about the fatigue in the work 
environment and how we wanted to get together with the agency 
to address this.
    And that was one of our biggest reasons to get back to the 
table and get a fair collective bargaining agreement was 
because we saw these new hires come in with a 30-percent 
reduction in pay and working at these busy facilities in these 
high-cost-of-living areas and not able to make ends meet, and 
they were waiting tables and doing anything they possibly can 
holding down as many jobs as they possibly could.
    The Chairman. Well, we have a problem here, gentlemen. Dr. 
Belenky, maybe you'd care to comment. Can these two things 
coexist side by side and have us talk about maximizing safety?
    Dr. Belenky. They can, but in a rather roundabout way.
    Performance is dependent on sleep in 24 hours, the total 
hours of sleep, however you split it. Divided sleep is good, 
can be fine, if it sums to 7-8 hours in 24 hours. If total 
sleep in 24 hours is adequate, then you probably need not to be 
so concerned about commute time or second jobs. It is when 
these cut into the sleep time that there is a problem.
    There are ways of directly measuring on--activity monitors 
that you could actually track people's sleep-wake history over 
days, weeks, months, unobtrusively, and if the total sleep/24 
hours were adequate, again, that would probably be all right, 
but, again--Yes.
    The Chairman. With all due respect, I mean, you're talking, 
I think, a little bit from a lab point of view.
    Dr. Belenky. Absolutely.
    The Chairman. And in the real world of being in a control 
tower, people aren't going to divide up their sleep very well, 
I wouldn't think. Maybe I'm wrong. Maybe both Randy and--I'm 
sorry--Director Babbitt and Mr. Rinaldi can comment on that. I 
mean, I think this is a very big issue.
    Mr. Rinaldi. Well, fatigue is real, as I said in my opening 
statement, and we've been wanting to address this for many 
years.
    And I applaud the Administrator for putting a workforce 
together, a task force to address it. We've come up with 12 
recommendations. We believe all 12 of these recommendations 
will help mitigate fatigue in the work environment.
    And, as I said earlier, the new collective bargaining 
agreement is fair, and it has gotten us back to--although not 
yet. In 2012, we will get back to the 2006 pay bands, which has 
taken some of the stress off the new air traffic controllers 
that don't have to have maybe two or three jobs anymore. We're 
not there yet. We are getting there.
    So there are a lot of things we're addressing there, but 
the 12 recommendations that are built on science--and it's not 
the union says or the FAA says--really is a conglomeration 
built on science, scientists from NASA, to say this will help 
fatigue in the work environment. That's one of the things that 
we're really pushing for.
    The Chairman. Work to be done. Thank you.
    Senator Cantwell. Senator Thune, do you have a second round 
of questioning?
    Senator Thune. Well, just a couple of things, Madam 
Chairman.
    And, Mr. Babbitt, following this string of sleeping-
controller incidents, Hank Krakowski, the CEO of the FAA's air 
traffic organization, accepted responsibility and resigned. 
That's a critical position, obviously, at the FAA. How long 
before you find a replacement, and what is the type of skill 
set that you are looking for?
    Mr. Babbitt. Yes, it was unfortunate. Mr. Krakowski was a 
professional, and I've known him for a long time. He had an 
excellent background and reputation.
    We are starting the process now. I couldn't tell you in 
exact terms. Sometimes some of the folks that you'd like to 
have might not be as interested in taking the job as we might 
want them to be.
    Certainly, this is a job that requires a lot of operational 
experience. This network is not unlike a large logistics 
network. This is a very complex operation.
    Just with respect to the operation itself, you've got over 
500 facilities that are manned with people on 24-hours--many of 
them, the vast majority of them are on 24-hour schedules. They 
have unique skills. They move. They have to be trained.
    The operational side of it, includes introduction to this 
system and obviously the new techniques coming with the NextGen 
procedures. How do we maximize those? How do we prioritize 
those?
    So we have, right now, a set of criteria that we're looking 
at. We're reviewing it within the Department of Transportation, 
and we're going to start our search very quickly.
    Now, the good news is, in the interim, David Grissell, who 
was Chief Counsel to the FAA is in place. David is a seasoned 
professional, 24 years at Continental Airlines. He has a lot of 
experience in big operations and saw a lot of transformation at 
Continental Airlines.
    He's familiar with networks, and I think he's doing a 
terrific job. He would rather go back to being Chief Counsel, I 
believe, than continue to be the COO. But he's doing a good 
job. In the interim, we've got a lot of good people in place. 
So my hope would be to have someone within the next few months, 
but it's hard to say when you're trying to recruit someone.
    Senator Thune. Accountability is important, but one 
individual is not solely responsible for these incidents. Have 
you made any other personnel changes that you believe will 
emphasize a change in approach throughout the rest of the 
management workforce?
    Mr. Babbitt. Yes, sir, we have. We have undertaken some 
pretty dramatic management changes. We've got about 10 
different areas where we have inside leadership changes, and, 
in some cases, people thought maybe it would be better to move 
on and do something else.
    We clearly have some cultural changes to make. One of the 
reasons that Paul and I were on the Call to Action was to 
clearly reinforce professionalism.
    As Senator Rockefeller mentioned, we can't regulate this. I 
can't regulate professionalism. I wish I could, but I can't. 
The vast majority of the air traffic controllers are very proud 
of what they do. They have great respect for what they do, and 
we've called upon them to help mentor people.
    Sometimes you see someone doing something less than 
professional. If you do speak up. It's your profession. We've 
really carried that message to them, that they need to be 
helping us police the professionalism.
    Someone can have 16 hours of rest, in terms of what Dr. 
Belenky thinks is adequate rest, but what if we found out that 
he played 36 holes of golf that afternoon? That's not 
professional. I don't care how much time off you had, if you 
didn't use that time wisely and take advantage of your sleep 
opportunity--it's not professional.
    I'm very pleased the professional standards group that is 
being built within NATCA is addressing these issues, because on 
some of this stuff you can't do top down. You have to have it 
come from the bottom up. They're inspired to do it. It's a 
proud profession. They're not happier about this blemish than 
anybody else.
    Senator Thune. Right. Let me ask just a general question, 
too, because you've implemented the nine-hour rest period. I 
think the IG had recommended 10. NATCA recommended nine-hour 
rest between evening and dayshifts only. Is that satisfactory? 
Does that rest period--is it something that you feel will be 
the most effective in mitigating fatigue?
    Mr. Rinaldi. One of the 12 recommendations was the nine-
hour break between the evening shift and going to what we would 
call the quick turn to the dayshift. That was backed with 
science and said that would give us an extra hour of sleep in 
our sleep bank, so to speak, as we were rotating through our 
shifts. That was backed with science and that was one that we 
recommended.
    The extra hour between--and I defer to the scientists 
here--the extra hour between the dayshift and the midnight 
shift, because you're starting your shift later in that 
deprivation period of midnight to six o'clock in the morning 
when your circadian rhythms are expecting you to sleep--you 
know, you're really focused on falling asleep at that point--
that we don't support it.
    We're working with the administration to show if science 
supports mitigating the fatigue, we're 100 percent on board. If 
10 hours is better than nine--Right now, it shows that 9 hours 
has the most benefit in between shifts than 10 hours, and if 
there is--nine hours are supported from a dayshift to a 
midnight shift with science, we will be 100 percent behind 
that.
    Right now, it doesn't show that. It actually shows the 
opposite, because you're starting your midnight shift in an 
area where--and you're working more hours in that dangerous 
period.
    Dr. Belenky. I agree with Mr. Rinaldi. It is a very tricky 
issue, and it depends critically on the timing of the sleep 
opportunity. If you place the sleep opportunity, as Mr. Rinaldi 
indicated, in the early to mid evening, that is the forbidden 
zone for sleep. Your body temperature is rising. All your 
systems are telling you to be awake, stay awake, and be alert. 
It is very difficult to sleep during that period.
    Therefore, 9 hours off from 3 in the afternoon until 11 or 
12 in the evening is not going to help very much, because it's 
not going to be a useable sleep opportunity. In contrast, 9 
hours from midnight to 9 in the morning, is excellent. You will 
sleep well and be able to capitalize on that sleep opportunity. 
Therefore, it isn't just the duration of the opportunity, but 
the key is placement with respect to the circadian rhythm.
    Mr. Scovel. Mr. Thune, if I may.
    Senator Thune. Yes.
    Mr. Scovel. You referred to our recommendation, which dates 
back to 2009 as part of our report on the three Chicago 
facilities.
    At that time, we recommended 10 hours between shifts, and 
it was our understanding, at that time, that FAA was about to 
change its internal order to specify 10 hours, as opposed to 8 
hours between shifts, in the 2-2-1 rotation. In effect, we 
endorsed that change.
    Since then--and Mr. Babbitt and Mr. Rinaldi have both 
referred to the work group that has recommended a move to 9 
hours.
    Look, we're not wedded to 10 hours. We would gladly defer 
to medical science on this question, but we think that the 
agency would be well-served to be guided by the science when it 
comes to naps or rests during a controller's work shift as 
well.
    It will be cold comfort, Mr. Thune, for the family of a 
victim of an aircraft accident if it's determined that it was 
due to controller error and that the controller was fatigued at 
the time and had been deprived of opportunities for rest.
    Senator Thune. Yes. Thank you. Thank you, Madam Chairman.
    Senator Cantwell. Senator Lautenberg, do you have a second 
question--second round?
    Senator Lautenberg. Yes.
    Senator Cantwell. Without objection.
    Senator Lautenberg. And I want to ask Mr. Babbitt a 
question. In 2006, a former FAA Administrator informed me that 
Newark Liberty air traffic control needed at least 35 
controllers to move traffic safely.
    Now, I don't know whether it was intended to be full-
performance people, but, right now, there are only 26 certified 
controllers manning the tower with 8 trainees. And what's the 
FAA plan to address the need to keep our towers fully staffed 
with certified controllers?
    Mr. Babbitt. Yes, sir. I'm not sure--I don't have the 
numbers available to me, but my understanding today is that we 
do, in fact, try to staff to traffic. So traffic flows change 
sometimes, and, therefore, you might want to increase staffing 
some place. On the other hand, sometimes the traffic falls off 
some place and staffing may need to be reassessed.
    A good example recently would have been Cincinnati, where a 
merger forced a move in traffic to other areas. So traffic in 
another city went up dramatically and traffic in Cincinnati 
went down.
    It takes us a while to migrate the people back and forth. 
So, specifically, my understanding today with respect to Newark 
Liberty is that we have a floor of around 28 and a ceiling of 
about 38. I could get you the absolute staffing numbers that we 
have.
    [The information referred to follows:]

    Safety is the top priority of the Federal Aviation Administration 
as it manages America's National Airspace System (NAS). An important 
part of managing the NAS involves actively aligning controller 
resources with demand. The FAA ``staffs to traffic,'' matching the 
number of air traffic controllers at its facilities with traffic volume 
and workload. The FAA's staffing needs are dynamic due to the dynamic 
nature of the workload and traffic volume.
    Based on the 2011-2020 Controller Workforce Plan that incorporates 
changes in air traffic forecasts, controller retirements and staffing 
ranges, the authorized staffing range for Newark 29 to 36.
    As of June 18, 2011, Newark has 28 Certified Professional 
Controllers (CPC) and six trainees. Three of these trainees are 
currently being used operationally and two of the six are scheduled to 
become CPCs by the end of 2011.
    In addition, Newark has brought on board two CPC-In Training (CPC-
IT) transfers and one new hire in 2011. An additional CPC transfer is 
expected in September. Newark is scheduled to hire three additional 
CPC-IT transfers in 2012.

    Senator Lautenberg. Let me ask you this: Newark Liberty is 
a complex airport due to high volume of flights, congestion, 
New York-New Jersey airspace, constrained runways.
    Now, so what have we got to do to provide the numbers that 
we need for Newark when my understanding is that 75 percent of 
the trainees don't make it through the program? So when you 
have a dropout rate like that or an incomplete rate like that, 
what do we do to get Newark up to date?
    Mr. Babbitt. Yes, sir. I think I made an earlier 
observation that we had difficulty with the previous collective 
bargaining agreement with the air traffic controllers. We were 
unable to attract seasoned controllers into complex facilities.
    That has changed, and those numbers that you're looking at, 
those are old numbers. We've had dramatic improvement since 
then.
    Now, if we needed to fill spots in, we're able to. For 
example, at Newark, we would be able to advertise a position, 
and a seasoned controller might come from a smaller facility 
and very easily upgrade into Newark as opposed to a new hire.
    And the fact that we had to put new hires into some of 
those complex facilities led to an exceedingly high washout 
rate, which was unfortunate. But we've cured that today.
    Senator Lautenberg. We still have increased salary for 
high-cost areas?
    Mr. Babbitt. Yes. Yes, sir.
    Senator Lautenberg. Is that still in place?
    Mr. Babbitt. Yes, sir.
    Senator Lautenberg. So if someone shifts in from another 
less busy airport to become a fully trained controller at 
Newark, that means they automatically will get an increase in 
their salary.
    Mr. Babbitt. Well, if they had come from LaGuardia probably 
not.
    Senator Lautenberg. No.
    Mr. Babbitt. Within the metropolitan area it would be 
probably the same salary. I'd have to look in particular.
    [The information referred to follows:]

    When moving Certified Professional Controllers (CPCs), the pay 
setting guidance is outlined in the Collective Bargaining Agreement and 
Appendices between the National Air Traffic Controllers Association 
(NATCA) and the Federal Aviation Administration (FAA).
    When a CPC transfers to the same ATC level facility, their base pay 
will remain unchanged. For example, the above guidance applies upon CPC 
transfer from an ATC Level 10 facility like LaGuardia to the same ATC 
Level 10 facility like Newark.
    When an employee transfers to a CPC position at a higher ATC level 
facility, base pay is increased to the minimum of the new CPC pay band, 
or is increased by 6 percent (6 percent), whichever is greater, not to 
exceed the new band maximum.
    Note: One-half (1/2) of the increase is paid upon initial transfer 
to the new facility; the other one-half (1/2) is paid when fully 
certified in the new facility.

    Senator Lautenberg. No, we try to keep the bi-state wars--
Mr. Rinaldi, the House Republicans have threatened to cut back 
FAA funds to 2008 levels. What impact would these proposed cuts 
have on our ability to hire and fully train new air traffic 
controllers?
    Mr. Rinaldi. That would be a big concern of ours to go back 
to 2008 levels for the obvious reasons that, as Mr. Babbitt 
said, from 2006 to 2009, we lost somewhere between 4,500 air 
traffic controllers--around 5,000 air traffic controllers, and 
in the last 5-year period, they hired somewhere over 7,500 air 
traffic controllers.
    It takes 3 to 5 years to train somebody to become an air 
traffic controller, and that training does put a lot of stress 
on the program, and a lot of our facilities are above the 25-
percent optimal trainee to CPC, certified professional control 
level.
    So if we went back to 2008 numbers and we looked at not 
hiring and continuing hiring, the fear that we have is 
currently we have about another 4,000 air traffic controllers 
are ready to retire or eligible to retire and they will be 
retiring soon. We have another wave of retirement. We haven't 
caught up from the first wave of retirement that we experienced 
in 2006 to 2009.
    Senator Lautenberg. I close with thanks to all four of 
these people who do a terrific job, and their teams do a 
terrific job, but we're going to hound you to death to even 
make it better if we can possibly do it.
    Thanks very much.
    Senator Cantwell. Senator Warner.

                STATEMENT OF HON. MARK WARNER, 
                   U.S. SENATOR FROM VIRGINIA

    Senator Warner. Thank you, Madam Chairman. Thank you for 
holding this hearing.
    And I want to commend Mr. Rinaldi and Administrator Babbitt 
for some of the actions they're taking. I think we all were 
surprised by these incidents, but I'm glad to see you've been 
working further together.
    This was not an area that I had a lot of knowledge about, 
but I'll remember a meeting I had with Mr. Rinaldi back in 
2007, and I don't think I, in all my time in public service, 
had more of a frightening session kind of getting Air Traffic 
Control 101 in terms of the potential wave of retirees, the 
challenges of attracting new folks, the ability to get through 
training periods, the ability to attract people to stay in this 
profession, the antiquated equipment and the need to move to 
NextGen.
    And I want to, again, echo what Senator Lautenberg said. I 
think you all do a good job. We need to constantly be vigilant.
    But I guess I'd ask you first if there's anything else? I 
mean, the remarkable thing is a lot of the things he said in 
2007 have all kind of come to pass, and I don't think a lot of 
our folks around the country would know kind of how close to 
the edge because there had been, obviously, a massive 
transformation of the air traffic control system in the early 
1980s when a whole lot of new people were hired. They've run 
through their cycle.
    And I just wonder if you've got any other--I know I've 
missed the first round of questions and your first round of 
testimony, but if you've got any other, at least at this 
moment, advice or admonition, hopefully not any more 
predictions similar to what you made in 2007.
    Mr. Rinaldi. Thank you, Senator.
    Unfortunately, I think that there's a lag in the system, 
and I didn't mean to, in my opening statement, to say that 
we're not concerned about operational errors. We certainly are 
concerned about operational errors.
    What I was trying to refer to is in the 2006 to 2009, and 
even years before that, there was a culture within the FAA to 
cover it up, hide it, and management's incentives were tied to 
it. And we weren't getting the information to address the 
safety concerns in the system, and that's why I applaud the 
administrator for really putting a just culture in place to 
address the safety concerns of the system.
    Unfortunately, I think we're going to see an increase in 
operational errors, as we talked about TARP, of better 
reporting, open reporting, we're going to start to really see 
where there are possible implications of safety in the system. 
And I look forward to working with the administrator and 
working with all of you to say, yes, we have a concern here. We 
need to address it.
    So as operations grow, as fatigue is a real problem in our 
work environment, and we look forward to implementing those 12 
recommendations to mitigate as much as possible--and I don't 
know if you can ever eliminate fatigue in a high-stressed, 24-
hour-a-day profession like air traffic control, but you could 
certainly try to mitigate it to the point that our cognitive 
skills are not impaired, and that's really where we look to go.
    So that we're embarrassed by what has happened, and we're 
proud professionals and we don't like any of the nonsense that 
is going on, and we want to make sure it never happens again. 
And I think the first positive step is to address these 12 
recommendations and to really address the safety concerns with 
the operational errors.
    Senator Warner. And do you both feel that you've got now 
the transparency and the training processes in place to make 
sure that we don't have this kind of cliff effect of retirement 
that we've run into in these last few years?
    Mr. Babbitt. No, sir, senator. The retirement rate is 
literally half of what it was 3 years ago. We're down to a 
steady state. Any corporation, any business that's been in 
business long enough gets to the point where you've got 
retirements at five, six, seven percent per year. That's where 
we're living. That's as it should be. And we can train to that 
without any problem and not overburden the system.
    But we were overburdened. There's no question. We were 
training twice that many of controllers for three straight 
years. It put a huge burden on the system. You know, we have a 
finite number of training facilities. When you put 30- and 40-
percent trainees into a facility, who trains them? The other 
controllers train them. So it's a big burden on everyone.
    But we're now getting the number of trainees down. We're 
very comfortable in the 20- to 25-percent range, and that's 
where we are today. So I'm very comfortable there.
    Senator Warner. Well, again, I appreciate the collaboration 
both of you are going to work on.
    And, again, Mr. Rinaldi, I just wish that all those things 
you'd said hadn't--the rest of them don't all come true as 
well.
    Administrator Babbitt, this is going to come a little out 
of left field, so you may not have the answer, but hope you get 
back to me.
    One of the things I've been very interested in as well is 
making sure we get additional spectrum out into the 
marketplace, and a number of years back in part of the mobile 
satellite spectrum there was an award made to a company named 
LightSquared that potentially would be another mobile broadband 
competitor, and there were certain questions about interference 
with existing GPS systems.
    Some of these concerns seem to have been raised now five or 
six years after the grants had been made. And nobody likes to 
give up spectrum, but some of the folks that I've talked to say 
there are ways that we can make sure there is another viable 
broadband competitor out there and still make sure, as we move 
into NextGen, that there is appropriate GPS protection for--or 
that appropriate NextGen GPS services are used and they're not 
going to have an interference.
    And I would just hope, as the FCC moves through this 
process, that you'll participate and not just have the approach 
that says, OK. You had this spectrum for 6 years. Now, all of a 
sudden, we're going to say there's potential interference here. 
So----
    Mr. Babbitt. Yes, sir. We're actively engaged in working 
through trying to find a solution there.
    One of the problems, of course, was the original intent of 
LightSquared was to use satellite signals. Back in November, 
they began to come up with the idea that they could enhance the 
signal by boosting it on the ground with terrestrial antennas, 
that are very powerful, literally 100 times more powerful than 
the original satellites forecast from space.
    Now, there is a technical solution available, but I have to 
tell you it would not be without consequence. The technical 
solution we would have to have would require the equipment 
redesigned to filter the interference, certify the redesign and 
reinstall it. It has weight. There are consequences to putting 
new equipment in airplanes.
    To require that would probably take in the 5-year-at-best 
time range. And then you're talking about equipping airplanes 
that have been designed, for the last 15 years to accommodate 
ADSB, automatic dependent surveillance broadcast equipment. 
This is where the airplane takes its position and develops it 
from a satellite and then broadcasts it. We have about 5,000 
commercial airplanes that use that equipment today, and about 
140,000 general aviation airplanes.
    Senator Warner. Clearly, anything would have to----
    Mr. Babbitt. Right.
    Senator Warner.--require a transition period, but I would 
like to see--And my understanding is the FCC's going to come 
out with what some of those costs would be. I hope you will 
actively participate, because, at some point, we're going to 
have to weigh the policy implications----
    Mr. Babbitt. Yes, sir. Yes, sir. No, we certainly will.
    Senator Cantwell. Thank you.
    Senator Warner.--but I just want to, again, thank you both 
for what you've been working on.
    Senator Cantwell. Senator Klobuchar.

               STATEMENT OF HON. AMY KLOBUCHAR, 
                  U.S. SENATOR FROM MINNESOTA

    Senator Klobuchar. Thank you very much, Madam Chairman. I 
apologize for being late. I was chairing a judiciary hearing, 
and, now, I'm the unenviable position of being last, where I 
think it means all questions have been asked, but not by me.
    So I first wanted to thank all of you. I was especially 
impressed by your comments and exchange with Senator Warner, 
Mr. Rinaldi, about the pride in your profession and how these 
recent events have been so disappointing and the work that's 
being done to fix it.
    As you all know, our commercial aviation system carried 
nearly 800 million people last year, many of them through major 
air transport hubs like Minneapolis-St. Paul International 
Airport.
    And we know that our accident rate has gone down over the 
years, but there are still issues, we know, when we have some 
of our traffic controllers falling asleep. Fatigue and sleep 
deprivation among air traffic controllers is a serious issue 
and I appreciate the chairwoman for having this hearing and all 
of you for taking this on.
    I know, Mr. Babbitt, the FAA has recognized and addressed 
the issue of fatigue, and the new staffing guidelines require 
two controllers in towers during nightshifts and 9-hour windows 
between shifts.
    Have you been able to quantify the effect of these new 
staffing policies on operational errors and runway incursions 
since they were implemented or is it too early?
    Mr. Babbitt. I believe, Senator, it's probably a little too 
early. Part of the problem is these towers--or whatever the 
facilities were, 27 towers, three TRACONs--that had a single 
person manning it.
    The reason they had such low staffing is there is very 
little traffic operations at those hours. So we wouldn't expect 
there to be much in the way of operational errors there, but it 
is too early to tell.
    Senator Klobuchar. Well, I know there were a spate of 
reports sort of in one time period about people falling asleep, 
but there seems like there haven't been, at least in the last 
month or two, since Secretary LaHood and others came out. Do 
you have any information on that?
    Mr. Babbitt. Well, we instituted a number of changes. You 
did miss sort of our review, putting 9 hours----
    Senator Klobuchar. No, I know the changes. I just wondered 
if you'd had reports of others falling asleep.
    Mr. Babbitt. Well, very candidly, I called for a top-to-
bottom review, and we did find some that happened earlier. Now, 
I'm sure if I go back further I'll find more. But we did find 
two that happened in January, both instances of people either 
observed with their eyes closed or observed sleeping. Neither 
of these were good, one in Los Angeles and one in Fort Worth. 
We're just adding that to the statistics.
    Senator Klobuchar. And I know when you visited me last 
week, we talked about this, but the rest periods between shifts 
and double staffing may require additional air traffic 
controllers. Does the FAA have an estimate of how these new 
policies could affect the demand for controllers?
    Mr. Babbitt. Well, in this case, we're talking about 30 
total out of 15,000. Now, that has a consequence, but we're 
also looking and working with the controllers to provide the 
same effective result as having two people together in a tower.
    As an example, we have facilities, where we have someone in 
a TRACON or a radar facility downstairs and someone in the 
tower upstairs, each of them alone.
    Well, we can put a console upstairs and let the person work 
radar up there. It's dark at night anyway, so they could work 
the facility and now there are two people at no cost, other 
than the one-time installation. So we're looking at things like 
that.
    Senator Klobuchar. So you'd have someone else in there, but 
they're working----
    Mr. Babbitt. Yes, they would be doing the same work that 
we're currently doing in a different location. We'll just put 
them in the same spot----
    Senator Klobuchar. I get it.
    Mr. Babbitt. Thus saving us the extra person.
    Senator Klobuchar. Yes. And then one last question: This 
spring, the Fatigue Work Group, composed of representatives 
from the National Air Traffic Controllers Association and the 
FAA, made a set of 12 recommendations that, if implemented, 
would address and mitigate the issue of fatigue among air 
traffic controllers.
    Do you know the status of review for this proposal? And do 
you expect to implement the recommendations? Are there other 
recommendations beyond double staffing during nightshifts and 
longer breaks between shifts that could be implemented?
    Mr. Babbitt. Yes, there's a number, and we have already 
implemented several of the pieces that came right out of the 
fatigue study.
    We are in discussions right now with NATCA. We're going to 
review and see what makes the most sense. In addition to the 
FAA and NATCA, we had some human-factors folks and people that 
introduced medical science to help us better understand fatigue 
and better understand how to mitigate it, so all of those 
factors are going to be in review.
    Some of them will require memorandums of understanding or 
modifications to the collective-bargaining agreement if we need 
to increase shifts or do other things like that. But we have 
had excellent cooperation between FAA and NATCA going forward 
on this.
    Senator Klobuchar. OK. Thank you.
    Anyone want to add anything?
    Mr. Rinaldi.
    Mr. Rinaldi. Thank you. The 12 recommendations, they're a 
good start, but they'll have to be implemented, tested and 
evaluated to see if we actually reach the goals which we're 
trying to do, and there might have to be more. The scientists 
and the workforce will continue to work together to see if we 
need more time in between shifts or whatever might happen and 
actually test controllers with wristbands and see exactly if 
we're really addressing the fatigue.
    More importantly--and Mr. Babbitt and Dr. Belenky touched 
on it--is the education factor about fatigue. We have about 
5,000 new air traffic controllers under the age of 30-something 
that, quite frankly, when we were all 30, we thought we were 
invincible. And we all were in college, we pulled our all-
nighters and did our tests and did everything.
    Well, we really need to make sure that we address fatigue 
and address it real and identify it and make sure that we do 
have proper periods to sleep and we use that to sleep and rest.
    Senator Klobuchar. OK. Very good. Well, thank you very 
much. And thank you, Madam Chairman, for having this hearing.
    Senator Cantwell. Thank you, Senator Klobuchar.
    And I think that does wrap up our hearing. We have 
discussed a variety of issues--training and the percentage of 
trainees that are acceptable, scheduling and changes to 
scheduling, getting more details on the operational errors and 
the meaning of those operational errors, and, obviously, the 
implementation of recommendations.
    So this committee is going to play an active role on 
oversight on all of those issues. I can't say that we're taking 
a legislative path at this moment, but I can tell you we're not 
ruling it out either. We are going to continue to be diligent 
until we feel that we have improved the safety and continue to 
implement those safety recommendations.
    So, thank you all very much for this hearing.
    [Whereupon, at 4:11 p.m., the hearing was adjourned.]

                            A P P E N D I X

Response to Written Questions Submitted by Hon. John D. Rockefeller IV 
                      to Hon. J. Randolph Babbitt

    Question 1. Administrator Babbitt, as you know, this committee has 
jurisdiction not only over aviation, but also telecommunications 
policy. Improving air traffic control safety and extending broadband to 
consumers are both priorities of mine. Over the past few months, there 
has been a lot of discussion about a decision coming out of the Federal 
Communications Commission (FCC) to allow certain satellite spectrum to 
be used for ground-based wireless broadband. I've heard conflicting 
reports on whether this FCC decision could result in networks that 
cause interference to the Global Positioning System (GPS), including 
those that aviation systems rely upon. I understand that the FAA is 
currently conducting tests to see if there is a technical solution to 
these concerns. Do you think that there is a possibility of a technical 
solution?
    Answer. The LightSquared signal design consists of two broad 
channels, an upper and a lower. LightSquared's plan has been to deploy 
the upper channel first, so that is what we have concentrated on for 
test and analysis. This upper channel would cause unacceptable 
interference and loss of GPS service to the existing aviation GPS 
receivers throughout a significant portion of the National Airspace 
System. This interference is caused by the overwhelming power 
difference between the LightSquared signals and GPS. The only technical 
solutions are to significantly reduce the LightSquared signal power, or 
to try to develop new GPS receiver technology that could coexist with 
the LightSquared signals. The engineering to develop a GPS receiver 
solution is technically risky and will be expensive, and deploying it 
would likely require replacement of all existing aviation antennas and 
receivers. The FAA estimates that such an effort--including design, 
standardization, development, certification and retrofit--would take at 
minimum of 7-10 years and require the modification of thousands of 
international as well as all domestic aircraft.
    On June 20, 2011, LightSquared released a press release identifying 
signal changes for their initial deployment plans, to use the lower 
channel first. We have conducted some preliminary testing and analysis 
that indicates that aviation receivers may be compatible with this 
lower channel. However, that same preliminary testing does show many 
other GPS receivers could be significantly impacted. Further study is 
required, and unless the upper channel is never transmitted this would 
only be a temporary solution.

    Question 2. Do you think that this could adversely impact the Next 
Generation Air Traffic Control System (NextGen)?
    Answer. If LightSquared were to use their upper channel, the effect 
on NextGen would be significant. Based upon aviation interference that 
occurred in testing and is supported by analysis, there would be a loss 
of a number of integral NextGen capabilities, including a loss of Area 
Navigation and Required Navigation Performance (RNAV/RNP), loss of 
Localizer Performance with Vertical Guidance (LPV) approaches, loss of 
Automatic Dependent Surveillance-Broadcast (ADS-B) services, and 
Cockpit Display of Traffic Information (CDTI). The interference would 
also cause loss of effective terrain awareness and warning system 
alerting, which could have a direct impact on safety since pilots would 
no longer have the necessary input to avoid collisions with obstacles.

    Question 3. Has there ever been a comparable period in FAA history 
where there was such a spike in operational errors?
    Answer. The identification and reporting of operational errors has 
continually improved year after year. We are and will likely continue 
to experience an upward trend in the number of reports due to the 
increased emphasis on reporting, the establishment of a voluntary 
reporting program and the introduction of technology that automatically 
detects losses of separation.

    Question 4. What data is included in the FAA's official count of 
operational errors by controllers?
    Answer. The FAA's official count of operational errors includes all 
losses of separation attributed to the Air Traffic Control system where 
less than 90 percent of the required separation was maintained. Losses 
of separation where at least 90 percent of the required separation was 
maintained are classified as proximity events. These counts do not 
include employee identified potential losses of separation reported 
confidentially inside the Air Traffic Safety Action program (ATSAP).

    Question 5. Have all of those losses of separation errors been 
included in the official count of operational errors?
    Answer. All losses of separation are not classified as operational 
errors; e.g., losses of separation attributed to pilots are classified 
as pilot deviations. As described above, proximity events and potential 
losses of separation reported confidentially to ATSAP are not included 
in the official count. The FAA's official count of operational errors 
does not include incidents that have been investigated and determined 
to be proximity events nor does it include incidents where the culpable 
party is not an air traffic controller.

    Question 6. How many of what the FAA has defined as the more severe 
errors (Category A and B) are reported only via ATSAP?
    Answer. Incidents reported only via ATSAP are held confidentially 
and are not categorized in the same manner as Operational Errors. 
Incidents reported to ATSAP are reviewed by a three-party committee of 
ATO management, Air Traffic Oversight and the National Air Traffic 
Controllers Association. This committee reviews each employee report, 
collects the information in a database and takes corrective action 
based on individual incidents and accumulated data.

    Question 7. What number and what percentage of total facilities 
have TARP totally operational?
    Answer. TARP is operational at all terminal RADAR facilities, but 
currently, is only turned on for one hour, twice a month for auditing 
purposes. With the finalization of our new Quality Assurance/Quality 
Control processes, it will be used 24/7, and full utilization will 
occur later this fall.

    Question 8. Of the total errors reported in the official count, how 
many were reported through TARP?
    Answer. Operational Errors identified via TARP, which is a 
technology that provides information in addition to operational errors, 
are not tracked separately from the standard official reporting system.
                                 ______
                                 
   Response to Written Questions Submitted by Hon. Maria Cantwell to 
                        Hon. J. Randolph Babbitt

    Question 1. How is the FAA doing so far this fiscal year with 
respect to operational errors? We are over halfway through the Fiscal 
Year. Is the trend for overall operational errors looking better or 
worse than for FY 10? Is the trend for the category A and B operational 
errors looking better or worse than for FY 10?
    Answer. For the current fiscal year through May 2011 the trend for 
operational errors is higher; operational errors have increased by 4 
percent over the comparable period in Fiscal Year 2010. As with total 
operational errors the trend for category A and B operational errors is 
higher; category A and B operational errors have increased by 1.7 
percent over the comparable period in Fiscal Year 2010.

    Question 2. It is my understanding that there is a letter grade 
given to each operational error. What is the methodology the FAA uses 
for determining whether an operational error is an A, B, C, or 
proximity error? What are the variables the FAA examines? Which 
variables are objective and which variables are subjective?
    Answer. Operational errors are categorized as A, B or C using the 
criteria below. Proximity events are not considered operational errors. 
The only criteria used to determine the categorization of airborne 
operational errors is the airborne separation values. These values are 
objective as the data is collected from the radar systems used to track 
the aircraft.
    Category A: Consists of a loss of airborne separation where the 
separation conformance is less than 34 percent. In events involving 
wake turbulence, it includes incidents where the lateral separation 
retained is less than 70 percent.
    Category B: Consists of a loss of airborne separation where the 
separation conformance is more than 34 percent, but less than 75 
percent. In events involving wake turbulence, it includes incidents 
where the lateral separation retained is equal to or greater than 70 
percent, but not more than 85 percent.
    Category C: Consists of a loss of airborne separation where the 
separation conformance is 75 or more, but the horizontal and vertical 
separation retained is less than 90 percent of the required separation. 
In events with wake turbulence where the lateral separation retained is 
equal to or greater than 85 percent, but less than 100 percent.
    See the attached Separation Conformance Review chart for a graphic 
depiction of the different categories.




    Question 3. Does the FAA investigate all of the operational errors 
that are designated category A and B? If not, approximately what 
percent of category A and B operational errors are investigated each 
year? In FY 10, how many category A and B operational errors were 
attributed to training incidents? What is the relationship between 
category A and B operational errors and air traffic controller 
experience levels?
    Answer. The FAA investigates all reported operational errors. In 
Fiscal Year 2010, there were 27 or 6 percent of the total category A 
and B operational errors (443) reported where on-the-job training (OJT) 
was identified as being in progress. The FAA completed a random 
sampling for experience levels with operational errors and identified 
that:

   Employees with greater then 15 years of experience had 34 
        percent of the reported errors.

   Employees with 5 to 15 years of experience had 30 percent of 
        the reported errors.

   Employees with less then 5 years of experience had 36 
        percent of the reported errors.

    Question 4. Does the FAA list causal factors for each operational 
error? Have any meta-trends been identified over the years? For 
example, is air traffic controller schedule and fatigue considered 
causal factors for operational errors?
    Answer. Yes, the FAA does list causal factors for each operational 
error.
    The Air Traffic Organization launched a safety management system 
(SMS) risk analysis process that brings together pilot and air traffic 
control analysts to examine significant events and determine what 
factors caused or contributed to them. This process has identified 
procedures and processes that need to be improved to reduce incidents, 
improve training and enhance safety. For example, we have recently 
identified the top five categories of events that contribute to risk in 
the National Airspace System. Through corrective action plans, we will 
be addressing the specific issues identified around aircraft arrival 
sequencing, unexpected aircraft ``go around'' procedures, compliance 
with altitude and other ATC clearances.
    Air traffic controller scheduling and fatigue are considered as 
potential causal factors for operational errors. The ATO, in 
collaboration with the National Air Traffic Controllers Association, 
has been researching the impacts of scheduling and fatigue on air 
traffic controller performance. The research has not revealed a direct 
causal link to operational errors at this time. The research has 
indicated potential impacts to employee performance and the ATO has 
made initial adjustments to controller schedules to address fatigue.

    Question 5. Are there proportionately more operational errors 
during the midnight shift than other air traffic controller shifts?
    Answer. No, there are not proportionately more operational errors 
during the midnight shift as compared to other air traffic controller 
shifts. For Fiscal Year 2010 there were 68 operational errors between 
the hours of 9:30 p.m. and 5:30 a.m. as compared to 1,819 for all other 
times.

    Question 6. Mr. Rinaldi attributes some of the increase in reported 
operational errors to the precision of the TARP software. He implied 
that the TARP software is not up and running 24/7 at all the facilities 
it is intended to operate at. What is the status of the TARP rollout? 
Should we expect there to be an increase in the number of operational 
errors as it is rolled out because of the precision of its electronic 
monitoring? If so, can you explain why?
    Answer. TARP is operational at all terminal RADAR facilities, but 
currently, is only turned on for 1 hour, twice a month for auditing 
purposes. With the finalization of the new Quality Assurance/Quality 
Control processes, it will be used 24/7, and full utilization will 
occur later this fall.
    Yes, the FAA expects there to be an increase in the number of 
operations where specific separation minimums are not maintained and 
are detected because the system is active on a 24/7 basis.
    Today's system predominantly relies on a human being recognizing 
when they have lost separation. It is very difficult for the human eye 
to recognize the difference between 3.0 nautical miles (required in 
some areas) vice 2.9 nautical miles (considered a loss). TARP is an 
automated computer measuring system that is able to precisely identify 
when specific separation minimums are lost.

    Question 7. My understanding is that prior to the recent incidents 
of air traffic controllers sleeping while on the midnight shift, air 
traffic controllers at some facilities were allowed to read books, do 
crossword puzzles, and do other similar low level activities while 
waiting for the scheduled flights to arrive. Are these types of 
activities now banned while the air traffic controller is working his 
or her position on the mid-night shift?
    Answer. Historically, activities, such as reading a book, have been 
permitted in our facilities during midnight shifts while waiting for 
scheduled flights to arrive. The recent incidents and the evolution of 
personal technology have caused us the review these practices. We are 
completing the coordination necessary to establish a national policy on 
these types of activities.

    Question 8. If an air traffic controller completes his or her shift 
and feels that they need a nap before they feel safe to drive home, can 
they currently use their break room to take a nap? Is there an overall 
FAA policy on this or is it facility-by-facility?
    Answer. Air traffic control facilities are places of government 
business and, generally, not an appropriate location for off-duty 
breaks and activities. However, safety of our employees is a priority. 
If an employee is too fatigued to safely drive home, we expect our 
supervisors and managers not to endanger our employees. Therefore, they 
should make arrangements for the employee to commute safely or get 
appropriate rest prior to departing.
                                 ______
                                 
      Response to Written Question Submitted by Hon. Tom Udall to 
                        Hon. J. Randolph Babbitt

    Question. Representative Shuster introduced an amendment to the FAA 
Reauthorization bill that would require in-depth analysis of 
rulemaking's impact on the economy, employment and private markets. The 
proposed amendment also would require the different safety rules for 
various components of the airline industry, such as passenger airlines, 
charter airlines and others. I would like to hear your thoughts on if 
this amendment would have any impact on the administrators' ability to 
nimbly address safety concerns such as those highlighted today?
    Answer. Many of the amendment's requirements are generally 
consistent with the requirements of Executive Order 13563, Improving 
Regulation and Regulatory Review. The FAA already incorporates its 
principles of economic impact analysis into our economic evaluation 
methodology. However, the additional depth of analysis at certain 
stages of rulemaking would mean that timelines would be impacted; an 
estimate is an average of 4 weeks additional time needed. It should be 
noted that while the amendment does require the FAA to ``analyze the 
different industry segments and tailor any regulations to the 
characteristics of each separate segment,'' this does not mean that the 
level of safety that the rule is intended to promulgate would differ 
for these segments. However, the economic analysis requirements are 
written in a way that could make it more difficult to quantify what we 
are required to analyze, which could impact the agency's ongoing effort 
to achieve one level of safety.
                                 ______
                                 
    Response to Written Questions Submitted by Hon. Mark Warner to 
                        Hon. J. Randolph Babbitt
    Question 1. What are the key concerns that FAA has regarding the 
possibility of spectrum interference resulting from LightSquared's 
proposed operations?
    Answer. The FAA's primary concern is with the upper channel that 
LightSquared has proposed. Government and industry testing and analysis 
have confirmed that there will be significant spectrum interference to 
GPS from LightSquared's proposed operations. The LightSquared signal 
design consists of two broad channels, an upper and a lower. 
LightSquared's plan has been to deploy the upper channel first, so that 
is what we have concentrated on for test and analysis. This upper 
channel would cause unacceptable interference and loss of GPS service 
to the existing aviation GPS receivers throughout a significant portion 
of the National Airspace System. This interference is caused by the 
overwhelming power difference between the LightSquared signals and GPS.
    Based upon aviation interference that occurred in testing, there 
would be a loss of a number of integral NextGen capabilities, including 
a loss of Area Navigation and Required Navigation Performance (RNAV/
RNP), loss of Localizer Performance with Vertical Guidance (LPV) 
approaches, loss of Automatic Dependent Surveillance-Broadcast (ADS-B) 
services, and Cockpit Display of Traffic Information (CDTI). The 
interference would also cause loss of effective terrain awareness and 
warning system alerting, which could have a direct impact on safety 
since pilots would no longer have the necessary input to avoid 
collisions with obstacles.
    A secondary concern is the compatibility between LightSquared's 
lower channel and GPS, although preliminary testing and analysis 
indicate that for aviation receivers this may be feasible though 
additional study is required. Another secondary concern is the 
potential to impact aeronautical satellite safety communications, as 
the spectrum available for these communications would be significantly 
diminished if LightSquared decides not to offer those services.

    Question 2. To what extent has FAA been working with FCC to resolve 
any concerns it may have with spectrum interference issues resulting 
from LightSquared's proposed operations?
    Answer. The FAA has been actively involved in tests to assess the 
potential LightSquared impacts. The FAA participated in the Government-
sponsored testing and analysis under the National Space-Based PNT 
Systems Engineering Forum (NPEF), and has also been participating as a 
member of the FCC-mandated Technical Working Group including, 
organizing and funding of the data collection for aeronautical GPS 
receivers. In addition, the FAA requested RTCA, Inc., to evaluate the 
potential impact to aviation receivers and provided funding and test 
results to support their report. The results from all of these efforts 
will be provided to the NTIA and FCC for their consideration.

    Question 3. Are you confident that the process and procedures put 
in place by the FCC will provide meaningful opportunity for the FAA to 
voice its concerns? Are you confident that the FAA's concerns can be 
successfully addressed through this process? If not, what else should 
the FCC or NTIA being doing?
    Answer. Per established procedures, FAA comments on the 
LightSquared proceeding are provided through the NTIA. FAA comments 
sent to NTIA are combined with comments from the other Federal 
agencies, and NTIA provides consolidated input to the FCC. The 
agencies' comments are considered as advice by NTIA, and may not be 
reflected in the final NTIA input to the FCC.

    Question 4. Which specific frequencies of spectrum where FAA 
operates has the potential of being negatively impacted by 
LightSquared's operations? To what degree are different types of 
operations relating to the FAA's role being affected? i.e., if 
activities in some portions of L-Band are of greater concern to the 
FAA, please indicate which portions.
    Answer. The primary concern is the impact of LightSquared's signals 
on the use of GPS, and all of the operations which GPS enables. 
However, LightSquared terrestrial operation in the 1525-1559 MHz band 
has the potential to also impact aeronautical satellite safety 
communications. Aeronautical satellite safety communications are 
standardized by ICAO for the 1545-1555 MHz band, and as a result are to 
be given priority access to that spectrum. While LightSquared has 
indicated informally that they will respect that priority within their 
system if they offer such satellite communications, it is clear that 
such provision would impact the operation of their terrestrial 
broadband network. As a result, the FAA is concerned that 
implementation of the LightSquared terrestrial network may result in 
LightSquared deciding not to offer such satellite safety communications 
in their network, and a de facto loss of that capability over the 
United States.

    Question 5. Understanding that these issues are still being worked 
out, are you able to walk us through the perceived next steps for the 
FAA and/or aviation receiver manufacturers with regard to technical and 
operational steps aimed at avoiding the interference, as well as 
possible mitigation strategies?
    Answer. It is premature to determine next steps. We first need to 
better understand the revised signal architecture that LightSquared 
would propose, and we will have to evaluate the compatibility of those 
signals with our GPS infrastructure. In order for LightSquared to use 
their upper channel in any capacity, we would have to develop new GPS 
receiver technology. The engineering to develop a GPS receiver solution 
is technically risky and will be expensive, and deploying it would 
likely require replacement of all existing aviation antennas and 
receivers. The FAA estimates that such an effort--including design, 
standardization, development, certification and retrofit--would take at 
minimum of 7-10 years.

    Question 6. Are there specific aviation components that you believe 
would have a higher degree of difficulty and require a longer lead time 
in terms of making the fixes that you anticipate may be deemed 
necessary after the release of the final report next week? How much 
time do you think you will need to resolve these concerns?
    Answer. Aviation equipage would take a minimum of 7 to 10 years. 
International and military equipage could take longer. FAA ground 
systems would take an estimated 5 years to modify. The potential 
modifications are not proven and there is risk of degraded performance 
for some applications as well as fuel and carbon penalties for the 
additional weight.

    Question 7. Has the FAA been in contact with LightSquared directly?
    Answer. The FAA has interfaced directly with LightSquared in many 
different forums. In RTCA we worked together to ensure all assumptions 
used for the analysis were accurate and complete. The FAA also was a 
member of the FCC-mandated TWG that was co-chaired by LightSquared. 
LightSquared has met with the FAA Administrator, and has met with 
different levels of management down to the engineering level. The FAA 
has participated in tests involving LightSquared technology. These 
tests were observed by LightSquared personnel who validated the test 
environment and parameters.

    Question 8. Does FAA have a comment on recommendations contained 
within the June 3 report from RTCA special committee 159?
    Answer. The FAA concurs with RTCA that the proposed LightSquared 
channel plans would cause unacceptable interference to aviation use of 
GPS. Regarding the definition of alternate channel plans involving 
signals farther away in frequency from the GPS band, from an aviation 
perspective such an approach should only be considered if it represents 
an end-state for LightSquared and is codified by FCC rules at current 
planned versus authorized power levels and preventing operation of the 
terrestrial component above the studied lower-band frequency. If 
alternatively the new LightSquared channel plan is envisioned as simply 
an interim step it would not be acceptable, since the final end-state 
configuration would have the same impacts on aviation as the currently-
defined LightSquared channel plans. Use of reduced power in the upper 
band is likewise not compatible with aviation use of GPS.
                                 ______
                                 
Response to Written Questions Submitted by Hon. Kay Bailey Hutchison to 
                        Hon. J. Randolph Babbitt

    Question 1. On several occasions the Department of Transportation 
(DOT), Inspector General (IG) has called for increasing the skill and 
knowledge requirements for applicants for air traffic control positions 
and transferring some of the expense for training to the applicant 
instead of the Federal Government. Many institutions are currently 
meeting those recommendations, but are not being utilized by the FAA.
    Additionally, in October, 2007, and as updated in February, 2008, 
the FAA established Human Resources Policy Bulletin #48. That bulletin 
identified graduates of CTO schools, as an additional hiring source for 
FAA to staff terminal control facilities. The bulletin also authorized 
special appointing authority for these individuals and authorized 
bypassing the FAA Academy for placement in terminal facilities. Why has 
the FAA continued to hire people without air traffic control training 
or qualifications from the general public when qualified applicants are 
available from the CTO schools?
    Answer. The Federal Aviation Administration (Agency) is committed 
to hiring the best qualified applicants to fill Air Traffic Control 
(ATC) vacancies. General public announcements applicants provide the 
Agency with a valued hiring pool comparable to CTO applicants. Like the 
applicants from the CTO schools, applicants from the general public 
announcements provide the Agency with a substantive, diverse, and 
qualified hiring pool. Among those candidates are individuals who 
possess backgrounds and abilities such as pilots, airport dispatchers, 
and military personnel with aviation training.

    Question 2. Why aren't CTO graduates being given priority 
designation over applicants without any air traffic experience?
    Answer. Trade schools, universities, or colleges that have CTO 
programs have been designated by the Agency as a Special Appointing 
Authority. Applicants completing the CTO process from a Special 
Appointing Authority have the opportunity to apply for ATC positions 
through Agency CTO vacancy announcements. These announcements are not 
accessible to non CTO applicants. Applicants from all hiring sources 
who are qualified are processed in accordance with the Uniform 
Guidelines on Employee Selection policy.

    Question 3. Why has the FAA not utilized or hired CTO graduates as 
authorized under this Policy Bulletin #48?
    Answer. The Agency values CTO graduates and the supporting 
institutions as a valuable hiring source. The current Terminal hiring 
pool contains 761 applicants. A total of 51applicants are identified as 
CTO graduates, which is 15 percent of the pool, and represents those 
graduates who are interested in a career with the Agency.

    Question 4. Why are these individuals required to attend the FAA 
Academy if assigned to a terminal facility when the policy bulletin 
indicates they are authorized to proceed directly to a terminal 
facility?
    Answer. Currently, all Agency ATC hiring sources are required to 
attend FAA Academy training. This requirement is necessary as the 
Agency needs to fill vacancies at medium to large Terminal facilities. 
The mandate to attend the FAA Academy equips the employees a greater 
opportunity for success considering the vacancies available, which is 
in the best interest of the applicant and the Agency.
                                 ______
                                 
Response to Written Questions Submitted by Hon. John D. Rockefeller IV 
                      to Hon. Calvin L. Scovel III

    Question 1. According to the September 30, 2010, DOT IG audit the 
new Air Traffic Control training program (ATCOTS) has been mismanaged 
and has had significant cost overruns. According to the report, the 
``FAA did not fully consider program requirements'' in designing the 
ATCOTS program, adding that ``it will be difficult for FAA to achieve 
the original ATCOTS program goals or any training innovations without 
significantly modifying the existing contract.'' How is the FAA 
modifying the ATCOTS contract to reach the goals of the contract?
    Answer. FAA has modified the contract to add training costs and 
hours. However, FAA's actions are not enough to ensure that ATCOTS 
goals will be attained or even to ensure that those controllers and 
facilities most in need receive timely training. This is because FAA 
continues to significantly underestimate controller training 
requirements. For example, for the third year of the contract, 
beginning September 2011, FAA estimates that $157 million is needed for 
ATCOTS support to meet existing training needs (not including 
addressing goals to improve and transform the training). However, FAA 
only plans to budget $93 million for the efforts.
    The mismatch between funding and training needs is causing delays 
in providing training, which is forcing many FAA facilities to 
compensate by conducting training with already-limited internal 
resources. Furthermore, FAA estimates that if ATCOTS spending continues 
at its current rate, it will run out of funds as early as June 2012, 15 
months ahead of the 5-year base period of performance. Therefore, the 
program needs to be rebaselined, and the contract should be modified 
accordingly. As part of these efforts, FAA should determine the best 
mechanisms to meet its goals for revamping how controllers are trained.

    Question 2. Am I correct in stating that one of the conclusions in 
that same report is that air traffic controller training may have 
contributed to current air safety problems? What information does the 
FAA have regarding the experience level of controllers who have been 
involved in an operational error or deviation?
    Answer. While we reported significant problems with the procurement 
and implementation of ATCOTS, we did not connect the training provided 
by its contractor under the ATCOTS contract with the recent increase in 
operational errors. Our objectives focused on the execution of the 
ATCOTS contract.
    There are several factors that could explain the rise in 
operational errors, including a higher number of developmental 
controllers monitoring traffic, increased voluntary reporting under the 
Air Traffic Safety Action Program (ATSAP), the operating environment at 
certain facilities, training issues (including ATCOTS), or a 
combination of reasons. The results of previous NTSB investigations of 
operational error incidents have not revealed a single ``silver 
bullet'' reason for why these errors occur.
    Currently, we are in the process of determining what type of 
information FAA has, if any, regarding the experience level of 
controllers who have been involved in operational errors or deviations. 
We also recently initiated an audit to evaluate FAA's process for 
tracking, reporting, and mitigating loss of separation events, 
including operational errors. As part of that audit, we are reviewing 
FAA's data on losses of separation.
                                 ______
                                 
   Response to Written Questions Submitted by Hon. Maria Cantwell to 
                       Hon. Calvin L. Scovel III

    Question 1. Are you satisfied that the FAA is accurately reporting 
all of its operational errors?
    Answer. We are not satisfied that FAA is accurately reporting all 
operational errors. FAA statistics indicate that operational errors 
have risen significantly over the past year. According to FAA data, the 
number of operational errors committed by air traffic controllers 
increased by 53 percent--from 1,234 to 1,887--between Fiscal Years 2009 
and 2010. However, it is not clear whether this reported increase is 
due to more errors being committed, to improved reporting practices, or 
a combination of factors.
    Historically, FAA's oversight of operational error self-reporting 
has been problematic. Since 2000, our work has repeatedly raised 
concerns that nearly 300 FAA terminal facilities relied solely on 
controllers to self report errors. In some cases, we found that the 
self-reporting process was subject to intentional manipulation. In 
2008, our investigations at the Dallas/Fort Worth Terminal Radar 
Approach Control facility found that air traffic managers at the 
facility intentionally misclassified 62 operational errors as either 
pilot deviations or ``non-events'' to reduce the number of errors 
reported at that location. Since that time, however, no evidence of 
this type of manipulation has been brought to our attention.
    Still, concerns remain about whether FAA is accurately counting the 
number of operational errors and sufficiently identifying the trends 
that contribute to them. For example, it is unclear how FAA's Air 
Traffic Safety Action Program (ATSAP) reports are factored into FAA's 
current counts of operational errors. Further concerns relate to FAA's 
recent implementation of the new metrics, which are designed to capture 
each incident where aircraft fly closer than separation standards 
permit. It is unclear how FAA will use these new metrics to assess 
operational error risks or improve safety. At the request of this 
Committee and others, we recently initiated two audits to assess FAA's 
implementation and oversight of ATSAP and evaluate FAA's process for 
tracking and reporting loss of separation events and efforts to analyze 
and mitigate identified risks.

    Question 2. Do you believe the FAA is doing a good job of 
understanding the root causes of its category A and B operational 
errors?
    Answer. At this time, we are not confident that FAA is identifying 
and addressing the root causes of Category A and B errors, which are 
the most severe. As we note in our prepared statement, there is 
considerable uncertainty about the causes of the recent increase in 
operational errors. Moreover, we have found that operational error 
reports are often inconsistently completed. For example, in our audit 
of three Chicago area air traffic facilities, we found that many 
reports did not provide complete information about the work schedules 
or other potential fatigue factors for controllers who caused errors. 
This does not give us confidence that FAA currently has all the data it 
needs for effective root cause analyses.
    We are in the process of reviewing how FAA determines the root 
causes of all types of losses of separation, including operational 
errors, and whether corrective actions are being taken to effectively 
mitigate their risk.
                                 ______
                                 
      Response to Written Question Submitted by Hon. Tom Udall to 
                       Hon. Calvin L. Scovel III

    Question. Representative Shuster introduced an amendment to the FAA 
Reauthorization bill that would require in-depth analysis of 
rulemaking's impact on the economy, employment and private markets. The 
proposed amendment also would require the different safety rules for 
various components of the airline industry, such as passenger airlines, 
charter airlines and others. I would like to hear your thoughts on if 
this amendment would have any impact on the administrators' ability to 
nimbly address safety concerns such as those highlighted today?
    Answer. We have not analyzed the proposed amendment and its impact 
on FAA's ability to address safety issues. However, safety is and must 
remain FAA's top priority.
                                 ______
                                 
Response to Written Questions Submitted by Hon. Kay Bailey Hutchison to 
                       Hon. Calvin L. Scovel III

    Question 1. According to FAA, there is currently approximately 25 
percent of the controller workforce in training. You indicate that the 
percentage is significantly higher at some of FAA's air traffic control 
facilities, such as 46 percent in training at Seattle TRACON. Is this a 
safety concern?
    Answer. This could certainly become a safety concern at specific 
facilities where the number of trainee controllers becomes large enough 
to affect the overall operations of the facility. While FAA has enough 
certified controllers in total, nationwide, many of the Nation's most 
critical, complex facilities have a very high percentage of controllers 
in training within individual facilities. For example, the Denver 
TRACON has 43 percent of its controller workforce in training, and the 
LaGuardia Air Traffic Control Tower has 39 percent. These percentages 
are bumping up to a level of concern--that is when almost half of the 
controllers at these facilities are in training. It is important for 
individual facilities to maintain a significant number of fully 
certified controllers in the workforce because only they can manage air 
traffic at all of their assigned positions while providing sufficient 
on-the-job training (OJT) for newly hired controllers. Nevertheless, 
the stress on veteran controllers increases as they train large numbers 
of new controllers while at the same time maintaining their own 
proficiency.

    Question 2. Are there any statistics that link the recent increase 
in operational errors to air traffic controller fatigue?
    Answer. We are unaware of any statistical links between controller 
fatigue and the recent increase in operational errors because FAA lacks 
sufficient data to make such a determination. FAA currently gathers 
only limited data on controller fatigue during investigations of 
operational errors. For example, FAA's operational error investigative 
reports include a section for information on the start and completion 
times of the current and prior shifts. However, this field may be left 
blank, and it is unclear whether FAA completes any analysis of these 
data to identify significant trends. In our 2009 audit report on 
controller fatigue, we recommended that FAA expand operational error 
investigatory requirements to include more detailed information on 
factors such as overtime, OJT, and work schedule that could create 
fatigue conditions to determine whether these factors are a 
contributory cause to operational errors.
    We also note that NTSB has warned about the problems of controller 
fatigue on several occasions, and how fatigue can result in operational 
errors. In its letter following the August 2006 Comair Lexington crash, 
NTSB cited four other incidents that provided ``clear evidence'' of 
fatigue. In each case, the controllers forgot critical information 
about the traffic situation and issued an inappropriate clearance as a 
result. The controllers compounded this error by inadequately 
monitoring runways and/or displays, thereby failing to recognize and 
correct the developing conflicts between aircraft. The following 
briefly summarizes fatigue factors present in the four incidents:

   March 23, 2006, Chicago--Runway Incursion. A controller 
        cleared an Airbus A320 to cross runway 4L and then cleared a 
        Boeing 737 to take off on the same runway. The pilots in the 
        departing 737 saw the A320 and stopped. The controller had 
        worked an 8-hour shift the previous day and was then off duty 
        for 9 hours. He slept only 4 hours before returning to work at 
        6:30 a.m.

   August 19, 2004, Los Angeles--Runway Incursion. A controller 
        cleared a Boeing 737 to take off at the same time that a Boeing 
        747 had been cleared to land on the same runway. The landing 
        pilots discontinued their approach. The controller had worked a 
        shift the previous evening until 11:30 p.m., then went home and 
        slept 5-6 hours before returning to work the incident shift at 
        7:30 a.m.

   September 25, 2001, Denver--Takeoff from Closed Runway. The 
        controller cleared a Boeing 757 to take off from a closed 
        runway. She had worked a shift at the tower from 5:30 a.m. 
        until 1:30 p.m. the day before the incident, and then had a 9-
        hour rest period during which she obtained between 60 and 90 
        minutes of sleep.

   July 8, 2001, Seattle--Runway Incursion. A controller 
        cleared a Boeing MD-80 to cross runway 34R at the same time a 
        Boeing 767 was on final approach to the same runway. The 767 
        applied max braking to avoid a collision. The controller was 
        working his third shift in 2 days, with an 8-hour rest period 
        between shifts.

    Question 3. Your recent audit on controller training metrics showed 
a training attrition rate of 21 percent for newly hired controllers in 
Fiscal Year 2009. Are there significant differences between this 
national rate and the rate of attrition at specific facilities?
    Answer. Yes, there are significant differences in the attrition 
rate of newly hired controllers at the national level when compared to 
the rate at some of the more complex air traffic control facilities. 
For example, our ongoing audit on staffing and training at FAA's most 
critical facilities found that the Denver and New York TRACONs, and the 
Newark Air Traffic Control Tower, had attrition rates of more than 70 
percent for newly hired controllers over the past 3 years. In addition, 
the Chicago and Houston TRACONs, as well as the Miami Tower, had a 100 
percent attrition rate for newly hired controllers in Fiscal Year 2009. 
Placing newly hired air traffic controllers at the most complex air 
traffic facilities severely limits their chances for successfully 
completing facility training. As a result, FAA needs to better 
scrutinize where it places newly hired controllers.

    Question 4. Your office completed an audit on potential controller 
fatigue issues in June 2009. What actions has FAA taken to address your 
recommendations?
    Answer. FAA agreed with our recommendations but has not yet 
implemented corrective actions. Specifically, we recommended that FAA 
approve recommended changes to FAA Order 7210.3 and implement the 
changes at all air traffic control facilities. The changes include: (a) 
increasing the minimum rest period between shifts from 8 to 10 hours, 
(b) increasing the time available for rest after working a midnight 
shift on the fifth day of a 6-day work week from 12 to 16 hours, and 
(c) allowing controllers to rest during their shift when not 
controlling traffic. FAA reported that it has formed a workgroup with 
NATCA to develop an overall fatigue management system that will 
identify and mitigate fatigue concerns for air traffic controllers. 
This workgroup recently developed 12 recommendations to address 
controller fatigue, including instituting a minimum of 9 hours between 
evening and day shifts. These recommendations were shared with the FAA 
Administrator and the President of NATCA on January 20, 2011. However, 
it is unclear when changes will be implemented. The FAA ATO is working 
with NATCA to conduct an initial assessment of the 12 recommendations 
to develop a Question and Answer product (estimated completion summer 
2011). This initial effort will be followed by an evaluation of 
implementation impacts and safety case validation. NATCA supports all 
12 recommendations and believes they can be implemented immediately.
                                 ______
                                 
   Response to Written Questions Submitted by Hon. Maria Cantwell to 
                              Paul Rinaldi

    Question 1. There is probably not nearly as much forced overtime as 
there was in prior years. When overtime is required, typically how does 
that get scheduled? Typically, is it an extra day or are extra hours 
tacked on to the existing workday? What is the best way of scheduling 
overtime to minimize the risk of increased air traffic controller 
fatigue for a 2-2-1 rotation?
    Answer. When overtime is scheduled, the assignment shift is posted 
on the schedule 28 days in advance, although changes may be made after 
the schedule has been posted if the need arises. However, if call-up 
overtime is necessary, an employee can be required to report to work on 
their regular day off with limited advance notice of the assignment. 
The nature of this unscheduled event can be fatiguing if the controller 
does not have time to plan for the overtime assignment. Additionally 
there is another category called ``holdover'' overtime. In this 
instance, the controller can be advised at the end of their shift that 
they will be required to stay for 2 hours of overtime or sometimes can 
be required to report to work up to 2 hours early.
    The Fatigue Workgroup did not model schedules with overtime. 
However, the Workgroup did find that all of the scientists indicated 
that any overtime, be it holdover or scheduled, increases the risk of 
fatigue. The best way to prevent overtime from contributing to fatigue 
is to properly staff field facilities so as to minimize the use of 
overtime. When overtime is necessary, it is least disruptive and poses 
the least risk to fatigue when it is known and scheduled in advance.

    Question 2. Why do you believe that the recent increase in 
reporting of operational errors is in part due to the certain strains 
on the system associated with high ratios of trainees to fully-
certified controllers?
    Answer. We believe that the recent increase is from the lack of 
experience of fully certified controllers compared to past years. As 
controllers gain experience the number of errors should decrease.
    High trainee ratios is an issue that NATCA has been warning about 
for over 5 years, and testified about before the Senate Subcommittee on 
Aviation Operations, Safety, and Security in March 2007, and before the 
House Subcommittee on Aviation in June 2008 and March 2007. Our message 
was consistent: The surge in new hires is placing a serious strain on 
the system and leading to safety concerns as experienced controllers 
retire and are replaced with trainees who require several years to 
become fully certified controllers.
    The strain comes in large part because high numbers of new hires 
require additional resources to train. NATCA testified before the House 
Aviation Subcommittee in May 2007 and again in June 2008 about the 
strains of hiring thousands of new controllers in a relatively short 
period (7,800 new hires over the past 5 years) would have on the ATC 
system. In the long-term, these new hires will enhance the safety and 
efficiency of our NAS, but in the short-term, this places a strain on 
facilities where they train because while achieving certification on 
position, trainees work under the direction of a fully certified 
controller or on-the-job-training instructor (OJTI).
    On the job training takes a toll on the instructing controller 
too--providing on-the-job-training to a new hire is extremely 
demanding, as the OJTI needs to be aware of every transmission and 
every keystroke the trainee makes. During OJT, a trainee works live air 
traffic, while the OJTI monitors both the trainee's actions and the 
radar or runway environment. The OJTI is held responsible for any 
errors made by the trainee. This essential training process increases 
workload for the OJTI and contributes to fatigue, particularly when 
these controllers are expected to train on nearly a daily basis.
                                 ______
                                 
     Response to Written Questions Submitted by Hon. Tom Udall to 
                              Paul Rinaldi

    Question 1. Mr. Rinaldi, you stated that many of the ATC employees 
have multiple jobs. I am concerned that economic conditions may 
contribute to more and more employees working other jobs besides their 
ATC job, which could prevent them from getting adequate sleep and make 
it more likely for them to make an operational error. What percentage 
of ATC employees have second or third job?
    Answer. During the period of imposed work rules (2006-2009), 
reductions in pay bands and a 30 percent reduction in salary for 
incoming controllers, as well as lack of compensation to those in the 
Academy training program led to many new controllers taking second or 
even third jobs to supplement their income. To the best of NATCA's 
knowledge, the need for supplemental jobs ceased to be a problem with 
the implementation of the 2009 collective bargaining agreement, which 
restores sufficient compensation for air traffic controllers for their 
work and should reduce the need for supplemental income.

    Question 2. Does the FAA take second or third jobs into 
consideration when determining the ATC shift schedule?
    Answer. The FAA does not consider second or third jobs when 
determining the controllers' schedule.

    Question 3. Is there any connection between the operational errors 
and shifts with at least one employee who has multiple jobs?
    Answer. I am not aware of any records that are available to answer 
this question.

    Question 4. What is the FAA doing to reduce fatigue concerns in 
controllers with multiple jobs?
    Answer. The FAA has worked collaboratively with NATCA on the 
Fatigue Mitigation, and has implemented several of the recommendations 
and is currently meeting to implement the remaining recommendations. 
However there are no specific guidelines for controllers with multiple 
jobs. The Agency has stated in the past it is the controller's 
responsibility to be fit for duty and the controller should manage 
their off duty time to ensure proper rest.
                                 ______
                                 
   Response to Written Questions Submitted by Hon. Maria Cantwell to 
                         Gregory Belenky, M.D.

    Question 1. In his written testimony, Administrator Babbitt says 
``the science of fatigue management for air traffic controllers is 
still an emerging discipline.'' Do you agree?
    Answer. Yes, it is an emerging discipline as is the discipline of 
fatigue-risk management, but we already know enough to know that one 
mitigation for on-shift fatigue and sleepiness is sanctioned, on-shift 
napping.

    Question 2. In your testimony, you speak about the impact of 
circadian rhythms on shift work. Is there one particular air traffic 
controller schedule that is better than others when it comes to 
minimizing fatigue?
    Answer. The best, most natural mitigation for fatigue is sleep. The 
problem with nightshift work is that night workers after getting off 
work are trying to sleep when their circadian rhythm is telling them 
that they should be awake. This leads to truncated sleep. Most 
nightshift workers are not able to sleep more than 5 hours off shift 
during the day. Again, the only way to ensure that nightshift workers 
are able to obtain adequate amounts of sleep in each 24 hours is 
sanctioned, on shift naps.

    Question 3. Is there a correlation between experience level and 
level of fatigue? In general, would you expect air traffic controllers 
working midnight shifts with lesser experience to experience greater 
fatigue than if more experienced air traffic controllers were working 
the shift? Of course, the complementary question is does coping with 
shift work and fatigue become increasingly difficult with age?
    Answer. The factors that interact to create fatigue are time awake, 
circadian rhythm phase, and workload. Being more experienced makes the 
work easier, thus lightening effective workload. Hence, being more 
experienced would lessen fatigue for equivalent work difficulty and 
duration. However, as you indicate, coping with shiftwork becomes 
increasingly difficult as we age.

    Question 4. One the factors impacting fatigue is time on task. For 
many air traffic controllers on the midnight shift, they do not get to 
rotate positions. I know that you recommend they be allowed to get 
recuperative rest during break times. Short of that, are there other 
steps based on the scientific literature that air traffic controllers 
can take to maintain a high level of situational awareness throughout 
their shift?
    Answer. In your question, you take away two good fatigue 
mitigators, on-shift napping and rotating positions. As they say in the 
British Army--``A change is a rest.'' The mitigations that are left 
are: (1) a late afternoon nap prior to going back on the nightshift to 
supplement the morning sleep obtained immediately after the previous 
nightshift, and (2) judicious use of caffeine on the nightshift. The 
less regularly you use caffeine, the more effective it is as a fatigue-
mitigation.

    Question 5. In your written testimony you describe in detail the 
Harvard Work Hours Health and Safety Group study of rates of medical 
errors associated with extended work hours and sleep loss. What should 
be the Committee's take-away from that?
    Answer. The Committee's take-away should be that an increase in 
sleep improves performance and decreases errors, incidents, and 
accidents in operational personnel working extended work hours. Sleep 
is the most powerful mitigator of fatigue in general and for fatigue on 
the nightshift.

                                  
