Remarks of Jim Hall
Chairman, National Transportation Safety Board
before the Aero Club of Washington
Washington, D.C., September 23, 1997

Thank you for inviting me.

When I became Chairman of the National Transportation Safety Board more than 3 years ago, I of course had no inkling about the level of activity my agency was about to enter. July 1994 saw the end of an unprecedented 27-month period in which the major U.S. scheduled airlines incurred no passenger fatalities: a billion passengers carried safely to their destinations.

I don’t think any of us are surprised that we didn’t read headlines about how safe the system was back then.

The USAir accident in Charlotte, North Carolina broke that string, followed by another USAir crash in Pittsburgh; and then crashes involving American Eagle in Roselawn, Indiana; another American Eagle near Raleigh-Durham; American Airlines in Colombia; ValuJet in the Everglades; Delta Air Lines in Pensacola; TWA flight 800; United Express in Quincy, Illinois; Comair in Monroe, Michigan; and Korean Air in Guam. And, we’ve lost two FedEx planes and had another major cargo accident in Miami.

This doesn’t count the Birgen Air and Aeroperu crashes that required our involvement in arranging deep-water searches for flight recorders, and assisting in those countries’ investigations.

The TWA flight 800 investigation is like no other in American history. I’ll talk about the technical aspects of it later in my remarks, but in the 14 months since that tragedy occurred, the Safety Board has had to seek a supplemental appropriation equal to more than half our annual budget. We have been on-scene on Long Island since the day of the crash, and, frankly, I cannot predict how much longer we’ll have to be up there, much less how long it will be before we can issue a final report.

A few years ago, most Americans didn’t know what the NTSB was or what it did. Today, unfortunately – and I repeat, unfortunately -- we find ourselves on the front page or the top item on the nightly news. A gas explosion in San Juan, a ship striking a mall in New Orleans, an Amtrak train derailment in Arizona, a safety study on passenger side air bag deployments – all news. NTSB investigates accidents in all modes. And still, we remain one of the federal government’s smallest agencies, and it costs each of you less to fund us for a year than to mail a post card.

Among all the organizations that ensure the safety of our transportation system, the Safety Board is unique. We focus on both the private sector and the government regulator, looking at every level of causality for things that can be fixed. We have no regulatory authority. We effect safety improvements only by indirect pressure. And, we know that our ability to be effective depends on our capacity to do quality work.

How can a small agency of 360 people with multi-modal responsibilities do quality work? One answer is leverage.

Long before this country began reinventing government, the Safety Board was leveraging resources by conducting its investigations through the party system. There is no way an agency of our size could maintain the expertise to conduct investigations of every type of aircraft model, every kind of engine, and all the systems of those aircraft, let alone all the trains, buses and cruise ships. We leverage our knowledge and our expertise with that of others. We call it the "party system." It has worked well and I’d like to see it continue to work well.

But it requires responsible commitment from all participants, and that is a continuing concern, as was demonstrated in the Guam investigation.

We understand and deal with party biases that are brought to the table. We know there are competitive pressures between the companies and we know the pressures generated by the contemplation of upcoming litigation. Nevertheless, the party system can provide a workable series of counterbalances.

In recent years, the American people have seen airline accident investigations that seemingly go on forever, and they might wonder, if we don’t know what caused the accidents, how can we be sure they won’t happen again, and soon? The short answer, of course, is that until we solve any accident, we’re not sure that it can’t happen again, and even then, we’ve seen accidents that look strangely familiar to earlier ones. But that is a very simplistic way to look at how accident investigation works.

Aircraft accident investigations are not like criminal investigations, where progress is measured only by an arrest. It is important that we determine the cause of accidents and, with a handful of exceptions, we always have. But more important to us is the safety recommendation, which is our most important product.

In many major investigations, we issue recommendations well before our final report because we see areas that need immediate improvement. And, if those recommendations are adopted then all of us have achieved our mission of improving transportation safety even before our final report is issued.

Following the crash of the DC-10 at Sioux City in 1989, we issued recommendations on 4 separate occasions before our final report was adopted, the first less than a month after the accident. Some of these recommendations dealt with inspecting fan disks and analyzing the adequacy of aircraft seat belts. Many safety improvements had been initiated before our finding of cause was ever issued.

Seven days after the crash of an ATR-72 in Roselawn, Indiana, we issued recommendations covering the operation of those aircraft in icing conditions. Twenty days after the ValuJet crash, we issued recommendations about the carriage of hazardous materials.

There is much we know about the circumstances of an accident before our final report, and that is no less true in the two current investigations that have engendered a sense of frustration in this country and in our organization – USAir flight 427 and TWA flight 800.

I want to talk a bit about these investigations, and focus on how the aviation industry has already changed for the better as a result of them, even though we haven’t yet issued final reports.

On September 8, 1994, USAir flight 427, a Boeing 737, crashed while on approach to Pittsburgh, Pennsylvania. After 80,000 hours having been spent by investigators, we hope to have a final report early next year, which would make this the longest investigation in our history. However, there is much we know about the circumstances of the accident, and much has been done to lessen the chances of a recurrence.

We have issued 20 recommendations so far as a result of this investigation. Here are some of the changes that have occurred in the industry since this accident:

• Boeing and the FAA are in the process of developing and certifying several modifications to the 737 main rudder power control unit (PCU) servo valve that will prevent the potential for reverse rudder operation.

• The FAA has proposed an Airworthiness Directive that would require the installation of newly designed PCUs within 2 years.

• Boeing is designing and certifying a new yaw damper system, which may be available on all 737 airplanes by late this year or early next.

• The FAA has issued a proposed rule to require modification of the input rod bearing on 737 PCUs.

• Most airlines are providing training to pilots on the recognition, prevention, and recovery of aircraft attitudes normally not associated with air carrier flight operations.

• Also, airline flight recorders will be upgraded so that existing aircraft will have at least 17 parameters and newly manufactured aircraft will have up to 88 parameters of information.

In addition, much has been learned about wake turbulence as a result of tests conducted during this investigation. This has been an exhaustive examination of the most popular airline model in the world.

Let me now turn to the crash of TWA flight 800 more than a year ago, which has resulted in the most extensive investigative effort in the Safety Board’s 30-year history, beginning with the incredible recovery of the victims and the wreckage from the bottom of the ocean.

• There were 677 surface-supplied dives,

• 3,667 scuba dives, and

• 209 remotely-operated vehicle dives.

• The diving and trawling took 10 months and covered 40 square miles of ocean floor.

• All 230 victims have been identified and returned to their loved ones, an almost inconceivable accomplishment, and one the American people should take pride in.

• We believe we have recovered between 95 and 98 percent of the wreckage of that 400,000 pound aircraft.

• The full-scale reconstruction of the main section of the fuselage is 94 feet long, the largest ever completed in the world. It consists of almost 900 pieces of wreckage.

What can we say about the progress of this investigation? I think we can say plenty. We know how the aircraft came apart, and what started the breakup sequence: the almost-empty center wing tank exploded. What triggered that explosion is not yet known, but we do know that the explosive fuel/air vapors in that tank set the stage for the catastrophic explosion.

A big issue is whether the precipitating event was a mechanical malfunction or an act of sabotage. That is why we have had two parallel investigations, ours and the FBI’s. Our finding so far is that there is no evidence of a bomb or a missile impact in the wreckage, but our investigation continues.

Yet, we believe that we have come up with recommendations that would make such an event less likely in the future. It was the explosive nature of the vapors in the fuel tank that allowed the blast – whatever its origin – to bring down flight 800. We believe that means exist that could, even in the short term, reduce the probability of a recurrence.

In December 1996, we issued four recommendations to the FAA that urged both short-term and long-term actions to reduce the potential for a fuel/air vapor explosion in the center tanks of Boeing 747s, as well as in fuel tanks of other aircraft. We suggested possible means to reduce the explosive potential of the fuel vapor, such as adding cold fuel to the center tank before takeoff, providing insulation or other methods to reduce the transfer of heat from the air conditioning units beneath the center tank, or inerting the tank by replacing the explosive vapor with a harmless gas.

We are not saying that these changes will prevent every accident in the future. We agree with the FAA and the industry that the policy of eliminating sources of ignition should continue. The problem is, there is no way we can assure ourselves that all ignition sources will ever be eliminated; TWA 800 shows that they haven’t been all eliminated yet.

It is no secret that the industry has not embraced our recommendations. But things have changed in the industry, and I think for the better, since the Board’s recommendations went out. I read with great interest the response of ATA, AEA, AAPA, AECMA and AIA – filed jointly – to FAA’s call for comments on proposals. While it disagrees with our recommendations, importantly, it contains two pages of "industry initiatives."

For example, industry plans to undertake a survey of aircraft or major fuel tank inspection programs to verify the integrity of wiring and grounding straps; the conditions of fuel pumps, fuel lines and fittings; and the electrical bonding on all equipment. This program will include not just Boeing 747s, but also Airbuses and aircraft of other manufacturers.

Folks are looking seriously at heat sources under center tanks. There is movement toward making Jet A fuel even less explosive. All to the good, if pursued with vigor. Next month, many of us will be at a conference convened by the FAA and the SAE here in Washington to discuss fuel flammability issues.

We have just completed a summer in which we conducted extensive flight tests in a Boeing 747 out of Kennedy Airport. Last month, we conducted a series of small explosive tests on the center wing fuel tank of another Boeing aircraft in England. We continue to conduct research at CalTech on Jet A fuel, as well as tests at Wright Patterson Air Force Base, and numerous other research facilities. We will be conducting fuel tank explosive tests in Denver next month, and are working with Sandia Labs in Albuquerque to develop computer models of explosions.

It is healthy that the industry engage in a thorough examination of the practicality and benefits of our safety recommendations. That is what is going on now with the center fuel tank recommendations. What I don’t want to see, though, is a reflexive action by industry that leads to our recommendations being rejected out of hand.

We were told shortly after the crash of TWA flight 800 that, even if the center wing tank exploded, it could not have brought down the plane. In addition, we were told that center fuel tanks of 747s are rarely in an explosive state. I think all of us are convinced otherwise by now. In fact, the industry submission I spoke of earlier indicates that the tanks are in an explosive state more often than not, although we are still analyzing our flight test data on this subject.

In 1991, when our engineers said that they believed the Lauda Airlines Boeing 767 had crashed because of an in-flight deployment of an engine reverser, we were told by both industry and government authorities that this was impossible. A few more weeks’ work convinced everyone of what had happened.

I am not claiming that the Safety Board is infallible. That’s certainly not the case. I am merely asking everyone involved in the regulation and operation of our airlines to consider the possibilities of things once thought not possible before deciding whether our recommendations have merit. We had no reason to think an unintended rudder failure could occur, much less lead to the crash of a Boeing 737; now, we’re not so sure. We once hoped we had eliminated ignition sources that could touch off vapors in an empty fuel tank, at least in aircraft like the 747; now, we know that is not the case.

I am often asked if it is frustrating to lead an organization that has no regulatory authority. Wouldn’t it be better if our recommendations were mandatory?

My answer is a simple "no." NTSB recommendations are adopted more than 80 percent of the time. We are effective because we are focused on cause and cure, nothing else. If we had regulatory authority, we would need to worry about cost benefit analyses, and court review, and a host of other problems. We don’t worry about these things, and I don’t think the American people want us to.

This is not to say that the job doesn’t have its frustrations. There have been some notorious cases of safety advancements that took many years longer than they had to. Post-accident alcohol testing in the railroad industry went into effect about a dozen years after our recommendations. Cockpit voice recorders in commuter aircraft took more than a decade. Bringing our flight data recorders up to the level of those in Europe has taken longer than I think is necessary.

A more recent point of frustration was one of the main issues coming out of the ValuJet accident. We investigated an accident 9 years ago in which undeclared hazardous materials started a fire in a Class D cargo hold of an American Airlines DC-9. Although the plane landed safely in Nashville, the fire had propagated to the point that the floor got soft above that cargo compartment and passengers sitting near it were moved away from the heat that was being generated.

The accident convinced the Board’s technical people of the inadequacy of the Class D compartment design concept, which was predicated on the fact that a fire would die out as soon as the oxygen was expended. When you have an oxydizer in the cargo hold, as we did on that American Airlines plane, and on ValuJet last May, the fire can continue to burn until it eventually breaches the protective liner. Despite the lesson learned in the Nashville accident, the FAA refused to require installation of either smoke detection or fire suppression equipment in Class D cargo compartments, citing cost/benefit analyses. Not surprisingly, following ValuJet, the smoke detection and fire suppression equipment overcame the cost/benefit analysis roadblock.

We cannot as a society afford not to do all we can to prevent catastrophic aviation accidents. The FAA has estimated that the direct costs of just one fatal airline accident – the Sioux City crash – totaled over $300 million, in addition to the enormity of the human tragedy involved. All of these things need to be part of the FAA’s calculation of new rules in the 1990s.

Let me close by saying that I am optimistic about the course of aviation safety. DOT Secretary Slater and Administrator Jane Garvey are bringing a new energy to safety regulation. There are tens of thousands of dedicated workers at the FAA and throughout industry who are faced with the challenge of aviation safety at a time of tremendous growth of our aviation system and changing public expectations about all aviation.

But this is a challenge worth all of our best efforts and I believe that the NTSB, through open and factual investigations, continues to make a difference. It is hard to quantify the accident that never happened because our recommendations prevented it, or the life that was not ended or forever altered. But perhaps one of you are here today because of the work of the NTSB over the past 30 years. Or, maybe more importantly, one of your loved ones is here.

Now, more than ever, the whole world is watching how we fulfill our responsibilities to the traveling public. Every precautionary landing is reported by USA Today, mishaps are covered live on CNN, and every major accident is the number one news story for a week. Yesterday, 35,000 flights carried their passengers safely to their destinations, twice as many as 10 years ago and half as many as 10 years from now. And there wasn’t one news story about any of those 35,000 flights, nor will there be, because that isn’t news.

Despite those remarkable safety statistics, if the already minuscule accident rate remains constant, Boeing predicts that there will be a loss of one transport category aircraft worldwide every week early in the next century. With those images on their TV screens, the American people won’t understand, accept, or believe that air travel is safe – nor should they. They expect – no, they demand – that we do everything within our power to make air travel as safe as we possibly can and we reduce the accident rate to as low as it can possibly go.

Our success depends on our ability to work together, to put aside our differences and our prejudices, to look for the best solutions, to implement them as quickly and efficiently as possible, and to make decisions based on what will make the system effective and safe. And, we must do it all under the increasingly bright glare of a very public and proactive spotlight.

I think I have one of the best jobs in Washington. I get to work in the transportation system to save lives. Thank you, as taxpayers, for giving me that opportunity, and thank you for listening to my remarks today.

 

Jim Hall's Speeches