FAA AIR TRAFFIC CONTROL COMPUTER MODERNIZATION STANFORD LIBRARIES TUESDAY, JUNE 16, 1981 HOUSE OF REPRESENTATIVES, COMMITTEE ON SCIENCE AND TECHNOLOGY, SUBCOMMITTEE ON TRANSPORTATION, Washington, D.C. The subcommittee met at 9:45 a.m. in room 2325 of the Rayburn House Office Building; Hon. Dan Glickman (chairman of the subcommittee) presiding. Present: Representatives Glickman, Harkin, Shamansky, Hall, Goldwater, Dunn, Carney, and Walker. Mr. GLICKMAN. The hearing will come to order. We have two congressional witnesses that are not here yet and we'll see if they're here. We'll wait a few minutes and if they're not here, we'll go ahead and proceed with Mr. Helms. I would like to call the hearing to order and indicate that this is the first of 3 days in what I consider to be perhaps the most significant set of issues facing aviation in America at this time. That's the issue of air traffic safety, the separation of air traffic, and the movement of traffic from one part of this country to another. This system is currently managed by a network of computers which the FAA, over the last several years, has indicated needs to be replaced because of being outmoded and because of increased air traffic. I and other members of the subcommittee have been concerned that the FAA might not be able to effectively manage its proposed new air traffic control computer program. Besides the huge pricetag on the FAA's recommended system of nearly $3 billion, I am also very concerned about their ability to manage such a large system acquisition. Their track record in the past has in some cases not been very good. Schedules have frequently slipped. Costs have exceeded estimates. And sometimes, research and development results have not met user requirements. I hope that these hearings will help clear up these and some of the other concerns of the subcommittee, which include: No. 1, whether the proposed computer system is worth the huge amount of money we would be putting into it; No. 2, whether the current system can be safely operated while a new system is being implemented; and No. 3, whether the American public is safe while we discuss the issue and even more importantly, whether the American public will be safe in the future as air space becomes an even more precious commodity. (1) In addition, I think it's obvious that during this particular time period, when the news reports are replete with statements made about proposed possible work slowdowns or shutdowns of the air traffic system because of disagreements between the FAA and their employees, that this set of hearings may focus on the long-term nature of what should the role of the computer be vis-a-vis the human beings that operate these computers. I think that maybe we in the committee can gain some good perspective. Today's hearings include Congressman John Burton, chairman of the appropriate Government Operations Subcommittee, which has jurisdiction over the FAA, as well as Congressman Bob Whittaker of my own State of Kansas, who has had an independent interest in the issues. In addition, we hear today from the new Administrator of the FAA, Lynn Helms, and tomorrow and Thursday, we hear from people with the General Accounting Office, from the Aerospace Industries Association, and from PATCO, the Professional Air Traffic Controllers Organization. Now I'd like to call on the ranking minority member, Barry Goldwater, for any comments he wishes to make. Mr. GOLDWATER. Thank you very much, Mr. Chairman. I welcome these hearings. The FAA operates one of the largest computer-based real time control system in the world. Part of this system is a 9020 computer, essentially the IBM 360. The computers operate in a multiprocessing mode with full redundant equipment available in case of equipment failure. Now continued significant growth is expected in our air traffic system. Especially in this growth in the future, we will see in general aviation and air taxi operations growing some 72 to 93 percent over the next decade. The new Administrator of the FAA, Mr. Lynn Helms, met with a number of us several months ago and said that he wanted to take an opportunity to look over this system at the FAA and these proposals. He told us that there were no sacred cows and that you can drag a dead skunk around long enough and it continues to increase in stinking. I think that the Administrator has probably had enough time to take a look at this system to find out whether we're on track or whether we're not. I think that that's the purpose of these hearings. The advanced computer program is currently in the program definition phase, so there is opportunity and time to make adjustments if, in fact, it is warranted. Therefore, Mr. Chairman, I think that these hearings are timely and appropriate to bring the Congress up-to-date as to where the FAA is, how they perceive this new system, and where we're going to go from here. Mr. GLICKMAN. Thank you, Mr. Goldwater. Now it's a pleasure to welcome our colleague, Mr. John Burton, from California, for hist statement. 1 STANFORD LIBRARIES 'ATEMENT OF HON. JOHN BURTON, A REPRESENTATIVE IN CONGRESS FROM THE 5TH DISTRICT OF THE STATE OF CALIFORNIA Mr. BURTON. Thank you very much, Mr. Chairman, members of e committee. Our subcommittee that's chaired by myself and the nking member, who is also a member of the Full Science and ech Committee, Bob Walker, unanimously issued a report called Air Traffic Safety Control Computer Failures," which I would like O submit for the record, along with a GAO letter, the subject of hich was the review of the Department of Transportation's reponse to the recommendations in the Senate report on the Federal viation Administration's en route air traffic control computer stem. As you are no doubt aware, the Senate Transportation Appropritions Subcommittee made very strong recommendations to the epartment of Transportation and the FAA aimed at certain nings. The GAO report tracking them say that very little, if nything, was done to adhere to what the Appropriations Commitee said. Basically, what our report found was that the FAA still has not ured the planning and management procurement problems that ave gone back over 10 years, that our subcommittee has held earings on that matter. The agency still lacks a unified, effective, high level planning nd management organization and a structure for dealing with utomation and data processing issues and needs. I kind of describe the FAA's procurement program as somebody coing out and buying 50,000 dollars' worth of furniture and then iguring out what kind of house they're going to build, rather than building the house first. That's what the Senate subcommittee ound-that they buy a lot of equipment, but they never have a mission in mind. They just keep buying the equipment and figuring some mission will turn up. We found that the failures in air traffic control computers are a threat to safety, despite statements by the FAA management. The FAA would say, and this was the old administrator, our great and good friend, Langhorne Bond, that if the automated computer goes out, they merely press a button and the new radar comes on board. Well, that's not quite the case. They would also say that the backup system is fine because they use it four times a week during the hours of 2 a.m. and 6 a.m. and they don't have any trouble. Well, the traffic between 2 and 6 a.m. is a little sparser than it is between, say, 7 and 10 or 4 and 7 in the afternoon. The backup radar was there because everybody knew the computers would be shutdown for maintenance and the backup radar was ready to go and there was no transition problem. During an actual flow of traffic in an emergency outage, we heard testimony that it took from 5 to 7 minutes to get the traffic back into a flow of control, and during that time, planes are lost from sector to sector and the chances of mid-air collisions are greatly increased. When the FAA could use such a statement-comparing 2 a.m. to 6 a.m. in the morning, a planned shutdown where the backup system was already in place to say that there's no problem when an emergency outage takes place, indicates to me that they were misleading the public. Only an idiot would try to think that the public would believe that the planned phasein at 2 and 6 a.m. was the same as an emergency computer outage. But then again, this was when Mr. Bond was the head of the Administration. The agency's computer performance and reliability data are not accurate or verifiable. The design of the reporting systems themselves do not require the right information to support a management information system that's sufficient to make important decisions. This means that despite the FAA's elaborate statistical presentations to the Congress, nobody really knows how reliable the computers and associated equipment are. Right now, the agency's outage reporting system only covers wholly unanticipated, sudden, total, and enduring failures. This does not accurately portray the safety-related problems faced by controllers; yet, today you're being called upon to make long-range multibillion dollar decisions based, in part, on this data. I'll give you three examples of how thin the ice is that we're really skating on, and these were cited in our report. An outage occurred during our investigation at the Freemont Center in Oakland in which both the primary and backup systems went out, and thank God, for a change, it wasn't foggy in the Bay area. There were serious safety implications. This was logged into the FAA's performance reporting system as a scheduled outage and won't show up in reliability figures. Figure that one out. One FAA region reported that it couldn't rectify its maintenance schedules with its maintenance logs for half of its equipment facilities. Another installation, we found, was officially listed as 96.8 percent reliable for the month of July in 1980, when it actually had been destroyed by a firebomb. But this did not violate the FAA's reporting rules. We found that you would have a breakdown in the computer and after, say, 2 or 3 minutes and the computer was out maybe for 3 hours, it suddenly became a scheduled breakdown for maintenance. You know, it was scheduled because it broke down. They played fast and loose with the figures. Bob Walker had tried to get Mr. Bond to come up with an accurate set of consistent definition and figures for breakdowns, for outages, for what have you, and we never got that. There's a different report or definition for an outage of 59 seconds than there is for one of 61 seconds. What difference that 2 seconds makes, I have yet to figure out. But if it's not out for a full minute, it's considered really nothing. I think that planes go fast enough these days that 30 seconds conceivably could lead to a midair collision. The FAA has not done an effective job of planning for remedying current deficiencies and ultimately replacing the existing system. The DARC system that was supposed to cure the current problems is marginal, at best. It's not a cure for the primary system. It's a little bit better than the old backup system but not much better. It was budgeted at $11 million; it's now up to $18 million and the agency's talking about almost $100 million to make it fully workable and redundant to the primary computer system. Without additions and improvements, it's maybe a step better than the oldtime STANFORD LIBRARIES ackup. With a lot of improvements, it's about 50 percent of the apacity of the present system. Under the Agency plan submitted to you, the DARC system is to e in use until the main computer is replaced in 1991 or 1992; yet, he agency has no real plans or budget for making DARC compatile with the collision avoidance system, which was supposed to be mplemented in the beginning of 1985. In short, the Agency has either planned to have two incompatile air traffic control systems in place at the same time for 6 years r really didn't believe it's own plans when it was submitting them o you. Or what is probably the case, they bought a lot of furniture nd then decided to figure out what kind of house they were going o build and they haven't figured that out yet. In the midst of all this disarray, the FAA is making staffing, udgetary and procurement decisions which often have 4-, 5-, or 6ear leadtime if they're to be carried out. It's important that the asis for those decisions are questioned and whether they can in act be implemented if they are correct. That was the purpose of the Senate Transportation Appropritions Subcommittee's very strong guidance to the FAA and as you vill see in the Controller General's report, they didn't pay much attention to the Senate subcommittee that gives them their money. I don't know how that will sit with that subcommittee this year. The second item I'd like to bring to the subcommittee's attention s that the FAA is or has apparently been unwilling to face up to and address these issues. As I'm sure you're aware, last year the appropriations committees of both Houses raised a number of these questions. They demanded detailed responses to a series of probems, many of which parallel those examined in our report. And to follow up, I made the request to GAO which I mentioned. The bottom line, if we can say, of that is: DOT has neither fully nor adequately addressed the recommendations of the In short, the agency was given direction in the committee report with threatened cutoff of funds to replace their computers and they failed to react. I would hope that the appropriations subcommittees this year and the authorizing committees forget about committee reports and throw it right into the money bill, that there be limitations in the appropriations. I came to Congress believing that no agency in government could be worse in procurement than the Pentagon. After having dealt with this agency for the past almost 5 years, I hope to God that the Pentagon is not worse than the FAA as far as procurement goes. We have found things that they have done as far as procurement that are some of the greatest wastes of taxpayers' funds you can find. Again, they see a nice toy and go out and buy it and they aren't quite sure what they're going to do with it because they haven't had the child yet to play with it. That was the whole direction of the appropriation committees' reports. It would be fair to say that the bulk of our report is aimed at the past administration of the FAA, and Mr. Helms, the new Administrator, can't bear the responsibility, I think, for too much |