I am pleased to be here this morning before such a distinguished international group. As a Member of the National Transportation Safety Board, I welcome this opportunity and appreciate your hospitality.

The aviation industry has seen tremendous growth around the world in the last decade. And no area has seen more of that growth than the Asian-Pacific region. The reality of this growth is that it unites us in our concern for aviation safety around the world. Aviation safety is an issue that we must all be dedicated to. In today's world, the impact of a major aviation accident no longer affects just the country and carrier that is represented on its tail. It impacts us all. U.S. citizens fly on virtually every foreign air carrier around the globe, as do many of the people from each of your countries. Aircraft parts are manufactured in a number of countries around the world. Air carriers fly into various airports around the globe and are serviced by local companies. All of these factors bring new challenges we must face in the next millenium as aviation industry continues to grow.

I have been asked to speak to you about maintenance human factors. This is a subject very near and dear to me.

When I arrived at the NTSB back in 1995, I asked staff to give me a list of accidents that had maintenance involvement, and I got this thousand yard stare, you know, the glassy-eyed one, "what are you talking about?" It seems that at that time, twenty eight years that the NTSB was in existence - nobody had ever tracked aircraft accidents with maintenance involvement. It took us approximately six months to sift through all the reports that had been written looking for maintenance issues and problems. We came up with slightly over thirty starting in 1961. After looking at the reports, I was left with many more questions. Often times the people conducting the investigation would come right to the verge of a maintenance issue, and then not ask the next question to explore the impact it had on the chain of events leading to the accident. It was no wonder that it took so long for our maintenance problems to come to the forefront. Nobody recognized the problems in the hangers or on the line.

Recently, things have changed a little bit. It started with the ValuJet incident in Atlanta, when we had a disc cause an uncontained engine failure in a DC-9 on takeoff, just prior to rotation and we lost the airplane - luckily with no loss of life. Five seconds later and that event could have had catastrophic results. When the NTSB conducted an investigation into the incident, we found a disc that had multiple corrosion pits in it. And from one of those corrosion pits - there were multiple pits - we found a crack that led to a failure of that disc.

But that's not all we found. Inside every one of those pits there was plating. Plating from the last shop visit when that disc went through the required inspections and the pits were not found. They plate after the inspection process. That flaw, which was known, and which was called out on paperwork as an inspection item, existed when that engine went through the shop. This was further complicated because it was overhauled in Turkey by a repair station that did not have the authority to do JT8D's for U. S. certified airlines at the time.

Shortly after that ValuJet adventure, a commuter airline crashed in Georgia. It lost a prop blade in flight causing a substantial imbalance, and fractured the gearbox. And as we often see, misfortune played a role. When the gearbox broke and twisted, it allowed the remaining prop blades to line up almost perfectly with the leading edge of the wing, disturbing the airflow and causing a substantial amount of loss of lift on one wing. The airplane made a little circle all the way to the ground. This accident was also caused by a maintenance misadventure. There is an Airworthiness Directive (AD) referring to this particular propeller which calls for repetitive on-wing inspections, non-destructive testing to look for the crack.

That crack was, in fact, detected by one of the airlines' maintenance people, and they removed the prop blade. They sent it back to the manufacturer's repair station, where another technician, who was "task-trained", unlike the first person, who was trained to level two in NDT procedures. This task-trained technician couldn't find the crack so he cleaned up inside the bore and put the prop blade back in service. The blade failed exactly where the AD said it would, within just a few cycles. This killed a bunch of people.

Then we come to the second ValuJet accident which occurred in the Everglades of Florida. Very painful for anyone with a maintenance background, for this was an accident that was 100 percent caused or initiated by a maintenance event. Only this one caused 110 people to lose their lives. I'm sure that everyone here has read about it, so I'll only briefly go over it. Basically, we had mechanics in the hangar of a Part 145 repair station, which makes it a little more difficult to control from an airline point of view, not follow the published procedures. They removed approximately 150 oxygen generators from an MD-80. When you read the paperwork, before you take them out of their little cocoon in the overhead, it provides instructions for the installation of a safety cap over the firing mechanism.

It's just a little piece of plastic, an inexpensive nickel and dime item, that prevents the hammer from striking the primer that would initiate the oxygen generator to start working. It's a chemical reaction and they get quite hot - something over 500 degrees just standing alone. Those mechanics in the shop, we believe there were four, possibly five of them, removed all of those oxygen generators and never installed any caps. They didn't have any caps. When they provisioned this aircraft for the check, they were never ordered. And the way the job card is written, you need to have the cap on the installed unit, it has to be left there until they're all installed and ready to go. And the unit removed needs to have a cap. For the life of me, I can't figure out why the manufacturers don't include an additional cap in the package when each one of these is shipped. It's just an insignificant cost item. Yet these units were removed, work progressed, the original caps ultimately were disposed of, and we had 150 or so of these generators sitting in cardboard boxes with no place to go.

Those of us who have worked in hangers know what it's like when an airplane leaves a check. The debris field left behind is extensive. Hangers can get to be messy. Sometime after that work, they had a hanger cleanup, and those oxygen generators were moved from the area around where the airplane was worked on into the stock room. We had a stock clerk, who wasn't trained in any hazardous material identification or shipping, put these units in the three cardboard boxes, tape them closed, and return them to ValuJet. They weren't labeled, the hazardous materials shipping documents were not properly filled out, and they ended up on the airplane.

When we performed some tests on the oxygen generators in the FAA's fire test facility in Atlantic City, virtually everybody was surprised with the degree of fire and the temperature that came with it. It was something over 3000 degrees. In fact the first time we tested it, the thermal indicator only went to 2700 and it pegged it out. So it's no wonder that airplane didn't survive with a Class D cargo compartment. For over ten years, the NTSB had recommended the installation of fire detection and suppression systems in Class D cargo compartments.

The sequence of events leading up to this accident, except for the last event where the ramp person threw the boxes on the airplane, was 100 percent maintenance. Every step of the way, every link in the chain, was maintenance.

These examples illustrate some of the problems we, in the United States, have experienced in the area of maintenance human factors in past few years. These problems are not isolated to our part of the world, however. The Asian-Pacific region has seen a tremendous increase in aviation travel over the past decade. Air travel in the region is expected to grow over six percent in the next twenty years.

I believe there are three key challenges we must address in the area of maintenance human factors. First, we need senior management to realize that training and program development in the area of maintenance yields a substantial return on investment. Second, aviation safety depends on effective communication among the entire "team," including the maintenance technician. And finally, as the aviation industry grows, we must address the retention and training of the maintenance workforce.

Initial efforts in the U.S. to focus on maintenance human factors issues and incorporate various programs that I was involved with, failed to recognize the significant role of senior management. It is this group of people who set the tone for the rest the organization and they must be able to see an actual return on the investments they are making. Senior management must understand the organization's problems and buy into the solutions in order for them to be successfully implemented. For a program such as Maintenance Resource Management (MRM) to really succeed, there must be quantifiable bottom line data to convince the financial specialists, not the aviation specialists. Companies that have addressed human factors issues are beginning to see real monetary savings in several areas. There has been a substantial reduction in re-work and less confusion at shift turnovers. Recently, a ten- day heavy maintenance was reduced by one full day, and that translates into about a $40,000 savings on one aircraft.

The challenge is to develop a tracking system that shows the dollar savings of these programs so that management can see the incentives of making the initial investment.

Studies of aviation maintenance organizations have shown that many aircraft mechanics don't like to ask for help or share their thoughts with others. That brings us to the second challenge we are facing in the area of maintenance human factors. One of the core elements of human factors is communication. Effective maintenance is predicated on a continuous flow of information. The information supporting maintenance must be timely, accurate, appropriate to the user, and in a form readily understood. Maintenance Resource Management (MRM) is one of the most effective tools in promoting communications within the maintenance department and between maintenance and flight operations. The objective of MRM is to reduce accidents and incidents, and improve performance. Improved performance includes such things as, eliminating maintenance delays, returns and diversions, makeovers and improper or incomplete procedures.

The key elements of successfully implementing MRM are:

• Management support. Management must be able to account for the returns on its investment. Training is expensive, and cost is often the main driver in deciding what programs and training to implement. Management must also believe that the programs they are implementing are effective and make a long term commitment to seeing them through.
• Training. The type of training your organization implements must be tailored to your specific needs.
• Feedback. Clear and open communication is essential for MRM to be effective. Words need to be followed by appropriate actions. The encouragement of feedback from the maintenance technician and follow up by management creates an atmosphere in which each individual is an integral part of the "team."
• Continuity. To succeed, MRM needs to be a continuous process that has ample visibility. Recurrent training is an important element of developing continuity.

The last challenge I would like to mention is perhaps one of the most serious ones we are facing in light of a growing aviation industry. We must do something in this industry to train and retain a qualified maintenance workforce. Aviation has seen tremendous growth in recent years, but unfortunately the number of qualified aviation mechanics has decreased. The quality of students and their basic technical knowledge is currently at a low point. This can be attributed to the fact that pay scales in aviation maintenance do not keep pace with those in similar businesses, for example the automotive business. Because of this discrepancy, the top students from aviation technical schools are not staying in the aviation business.

Also contributing to this decrease is the aging of the pool of licensed mechanics, the failure to attract a sufficient number of potential technicians into training and the reduced number of technicians being trained by the military services. Public perception and societal status of traditional blue-collar employment have also made this profession less attractive to many young people. I use the word "profession" intentionally because an aviation maintenance technician is a professional. There is a substantial amount of time and cost associated with this profession. An aviation mechanic has some 2200 hours invested in getting his certificate and earning the right to be tested on his knowledge. Unfortunately, this is where his training stops.

Training is essential across the board in aviation operations. The maintenance individual is a key component of this operation. Unfortunately, the training and curriculum for maintenance is long overdue for an upgrade. As aircraft are becoming more sophisticated and complex, it is more important than ever for the mechanics working them to be technically trained and qualified. The quality of the aviation maintenance professional has direct impact on aviation safety.

In closing, I would like to challenge each of you to become part of the effort to solve the human factors problems we face in the aviation industry. I believe the solution is multifaceted and requires a systemic approach. We have only recently convinced the industry to really focus on maintenance human factors, and I believe that we must be committed to addressing this issue head on if we are to secure the standards of aviation safety that the flying public deserves.