'Productivity and Safety Go Hand-in-Hand'

Interview with Prof. Gary Eiff, Purdue University, Dept. of Aviation Technology
A study showing a rising number of accidents, and a growing portion of them related to maintenance mistakes, is generating considerable attention within the industry and in the media (see ASW, May 3).

The study was directed by Prof. Gary Eiff at Purdue University. Eiff has years of experience researching aviation maintenance processes. The latest study was based on a line-by-line reading of some 1,300 accident and incident investigations conducted over a 20-year period by the National Transportation Safety Board (NTSB). The findings implicate maintenance problems far more than generally thought. Given the attention this study has generated, some salient questions were put to Prof. Eiff:

ASW: What did you think when you first saw the graph of rising accidents?

Eiff: I was surprised. I thought we'd see a slight increase in the number of accidents with more flying, but also that the number of accidents would have dropped a bit after the 9/11 attacks, because schedules were reduced and some older airplanes were laid up.

The system is not significantly less safe than, say, a year or two ago, but we have to get together as an industry to address the trend.

ASW: How do you explain the difference between your accident history and the NTSB table? Even though both data sets show a rising number of accidents, some significant differences emerge. For example, your table shows about 25 accidents for 2002, and the NTSB statistical database indicates 34 accidents.

Eiff: For that year, as an example, we only looked at closed cases. A number of investigations were still open. Our review was only of closed cases involving Part 121 operations.

ASW: You assert that most accidents involving mechanical failure are maintenance-related. Why do you believe this?

Eiff: Let me give an example - mechanical failure due to oil contamination. Something in the maintenance process caused the oil to become contaminated.

ASW: What trend do you see for maintenance-related accidents?

Eiff: I don't see a trend toward improvement. I think we're getting more real about what is occurring in the fleet and we're seeing what's been going on for a long time.

ASW: You found that some three-quarters of maintenance-related events involved a failure to follow procedures. What does that high fraction suggest?

Eiff: The industry has put a lot of time and effort into increasing compliance with procedures for pilots. We need to do that for mechanics.

Furthermore, many mechanics don't look at policies and procedures as an error defense, as a way to stay out of trouble. So there's an educational process here.

ASW: What can operators and maintenance facilities do to achieve greater compliance with procedures?

Eiff: There are known techniques for identifying at-risk behavior. At one carrier using our safety metric process, there was a 70 percent reduction in at-risk behavior in a six-month period.

In many instances, poor workflow exacerbates the problem. Our process-mapping approach can help streamline the operation. We have shown that you can get productivity and safety gains. They go hand-in-hand.

Also, the industry tends to promote excellent technicians in the belief that they'll be good managers. That's not always the case. We have a coaching process that improves their people-managing skills.

ASW: You suggested improved "process mapping" to address organizational deficiencies. Can you be more specific about what actions are involved?

Eiff: We look at every step, from input to output. We document the way the organization thinks a process is done, then we look at how it is actually done on the line. We look at what's necessary and what's not pertinent. We look at patches to the process, which tend to add delays and more potential for error. As an example, Australia announced that $10,000 fines would be imposed on the airlines for passengers arriving in Australia without required visas. So this one carrier with whom we were working had its people at the boarding gate checking for visas just before passengers walked into the jetway. The visas could have been examined at check-in, where the passports were examined. The gate check had no value added.

The same "process mapping" can be applied to maintenance.

ASW: What can or should the FAA do?

Eiff: The FAA struggles to execute Congressional mandates without the budgets to support them. Having said that, the FAA could promote research to make mechanics more productive and effective. That research money has dried up.

ASW: Is there a problem with clarity and ease of understanding of the maintenance manuals?

Eiff: Let me just say that we have some basic issues, such as which page is the most recent? In our work, we've found incorrect tasking orders and manuals, but they don't get changed because the process of changing them is so difficult.

Electronic manuals hosted at one site and updated every morning is one way to ensure that the documentation is current, and some carriers are starting to go electronic.

ASW: In one out of five events, inspections of maintenance work were inadequate or missing. What can be done to strengthen this final line of defense?

Eiff: The [low] level of required inspections is pretty disturbing. For example, the RII [required inspection item] list typically is pretty small. Obviously, some thought must be given to expandng the scope of required inspections.

ASW: Is the FAA adequately resourced in terms of people and money to provide maintenance oversight?

Eiff: I'm concerned. A lot of FAA people are well-intentioned, but they're caught in a Catch-22. They have tasks to do with inadequate time and resources. How can you have oversight when there isn't enough money to travel to a troubled operation? >> Eiff, e-mail geiff@purdue.edu <<


--------------------------------------------------------------------------------