Q&A: Plugging Holes in Transportation Safety

Catastrophic school bus accidents are rare events. But when they do occur, it’s important for the industry to find out what went wrong and how a similar tragedy can be prevented in the future.

The National Transportation Safety Board (NTSB) investigates these types of accidents, sending experts to the scene to analyze the facts and ultimately write reports and issue recommendations aimed at improving safety.

To provide insight on the workings of this federal agency, SBF Managing Editor Thomas McMahon spoke with Bruce Magladry, acting director of the NTSB’s Office of Highway Safety.

Magladry, a former police detective, has been with the safety board for about 18 years, serving as a human factors investigator in all modes of transportation before moving specifically into highway safety.

SBF: How does the NTSB decide which school bus-related crashes to investigate?
MAGLADRY: The NTSB investigates accidents in all modes — aviation, rail, marine, pipeline and, of course, highway. As the Office of Highway Safety, we’re the entity that’s responsible for investigating major highway accidents for the lessons that are learned.

Choosing which accidents to address is a rather daunting task. There are about 20,000 highway accidents a day in the U.S., and the Office of Highway Safety has a staff that’s historically about 30 people. So if you compare the numbers, you can see that we have to be very judicious in our selection of accidents. That selection is part science but part art as well.

One of the things we look at is the level of public interest in an accident event. In a school bus accident, that’s always very high because of the cargo that’s carried. We look at the circumstances as we get them from police to try to determine whether we can make a meaningful difference. We’re always looking for the national issue that we might be able find and address.

Since we’re addressing a relatively small number of accidents, we need to get the biggest bang for the buck. Is it an issue that we have already addressed? If we’ve already addressed it, we can’t afford the time or resources to re-address it. The staff is small, so that enters into my decision as well. We have a communications center here; it’s 24 hours a day. Those people search the Internet, they search the media, they find out about accidents. They make notifications to us of many more accidents than I can actually take. They’re the first level of filter.

For school bus accidents, I want to know about just about every one. I certainly want to know about the ones with an injury or a death. To give you a sense of scale, I want to know about tractor-trailer accidents when there are three or more dead. That doesn’t mean I’m going to investigate it; it just means I want to know about it. We’ll have one of our duty officers make some phone calls to the police and get some of the circumstances before we make a decision as to whether we want to investigate.

In a year, we’re only going to look at somewhere between 40 and 50 accidents in any depth, and we’re only going to look in great depth and write reports on somewhere between six and 10. So if you take those numbers and you look back at the 20,000 accidents a day, you realize that I’m doing something on a very small scale.

One of the constraints that the safety board works under is that we’re often making recommendations to people based on that single event. So the depth of our investigation has to be quite great. We’re not going to be able to convince anybody to do anything unless we’ve really done our investigation in a thorough manner. What we would like in the end when we finish gathering the facts and analyzing them, presenting them to recommendation recipients — like a school or a government — we would like them to say, “I see the problem. It’s obvious to me now.” And then they’re going to go ahead and fix it.

Since we have no regulatory authority and can’t force anyone to do anything, it’s the force of our argument — the logic of our argument — that is necessary to persuade people. So we do investigations in great depth.

Typically, how many investigators are sent to the site of an accident?
We have two teams, and I generally send six people. There is an investigator in charge — a senior investigator who coordinates everything. Then there is a mechanical engineer who is looking at the vehicles — how they operate, how they’re designed, how they’re maintained. I send a highway engineer. This is a civil engineer kind of person, and he works on things like roadway design, geometry of the roadway, the signage, the surface and coefficient of friction, the traffic volume, the speed, the accident history — everything that has to do with the infrastructure itself.

I send a human factors investigator, who is someone with a psychology background, looking at why people behaved in the way they did.

I send a survival factors expert. This is someone who has knowledge on injury and death mechanisms. So in a school bus accident, for example, we’re looking for how did the child get hurt? You can get bounced around; what is there to bounce around and hit against. What forces might be involved.

Then I also send a motor carrier expert. That’s a person who is knowledgeable on the operations of a transportation business. This is someone who goes and looks at the maintenance records, the training records, background of the driver, background of the company. And they are knowledgeable on the rules and regulations that affect school bus transportation. So, six people.

What do the investigations entail?
In any investigation, I generally send six people, but that’s not enough to ferret out all the information that you can gather in an investigation. So the NTSB has what we call a party process. We designate parties to an investigation when we come. These are organizations that had some material role in the accident. In a school bus accident, it would be the school district, the owner of the bus if it’s not the school district, the driver, the manufacturer of the bus, and it might also be the manufacturer of the transmission, brakes or engine if they’re different from the bus manufacturer.

It would also involve the police, the state Department of Transportation, the owner of the roadway — state, local. These are all people who have local knowledge of equipment, operations or infrastructure. We bring all those people together at the end of the first or second day, and we ask for their assistance. We ask them to actually supply us with people. For example, we’re going to want drivers’ records, so who keeps the records? Are the records electronic or paper? Are the employment records and the training records in a different place? With some legwork, we can find these things out. It’s a little easier if we have someone from the school board who knows where that stuff is and will gather all of it.

We’re really interested in discovering what occurred in an accident and to prevent it from occurring again. That’s a bit different of an approach from, say, the police investigation. Because police investigations are generally looking to place blame, fault, to issue a ticket, to arrest someone. We’re, in some sense, much less interested in the outcome of that kind of investigation than we are in where the holes are in the system that need to be plugged so it can be prevented the next time.

As the process goes along, we’re gathering facts — just the facts. We’ll check out things like how long the bus is and what the air pressure in the tires is until we have all the facts right.

{+PAGEBREAK+}

We do that for about a week. Maybe ten days. In some cases, it’s as long as two weeks. We’ll be on scene, where the accident occurred, gathering these sorts of facts. For example, if we have a brake issue in the accident, we could conceivably take the automatic slack adjusters off the brakes of the bus and physically take it to the manufacturer, in some other state, presumably, and we’d tear it down with their systems. Not let them tear it down and tell us what they saw, but assist us in tearing it down so that we’re overseeing the process.

Ultimately, each of the six investigators writes a factual report in their area of expertise. Before we’re satisfied with the facts, we send those reports back to the parties. If there are mistakes, then we’ll discuss those mistakes. And we’ll make changes as appropriate so that everybody is in agreement — the facts are the facts are the facts.

Once we get them through all the parties, we write analyses of these factual reports. Then we put each of those staff draft reports — both the factual and the analysis — together in a larger report, which is the safety board’s Report. Our analysis includes a probable cause of the accident — why it occurred — and recommendations to fix those problems. The recommendations are addressed to whomever is capable of making the fix. It’s often a government agency like NHTSA [National Highway Traffic Safety Administration] or the Federal Highway Administration or some state. Sometimes it’s to an individual company to make a design change. Sometimes it’s to associations.

Then, once the staff produces a final draft factual and analysis report, we go to the board — the board, which is the head of the agency, is composed of five members appointed by the president and confirmed by the U.S. Senate — and we have a large, open public meeting in which we discuss all the facts and analyses, pros and cons, and the staff recommendations. The board members then vote to adopt the report or conceivably to not adopt it. They have the ability to accept it, to not accept it or to change it. The staff is very good here, very professional, very expert. The number of changes that occur in that public meeting with the board are generally limited. Then it’s published, the recommendations go out and we hope that the recipients will follow our recommendations.

Are the recommendations typically heeded?
Historically, an average of 82 percent of our recommendations are indeed implemented. That’s over the course of our 30-year history. The number fluctuates by a percentage point or so, but that’s pretty solid. We think that that’s an exceedingly high rate, because we have to convince people to spend time and money based on the logic of our argument, as I stated earlier. We’re pretty excited about that. We wear that as a badge of honor.

Does the NTSB follow up on recommendations to see whether they’ve been implemented?
It is a big part of our business. There is a separate office that is tasked with that: the Office of Safety Recommendations. They send the original letters to recommendation recipients, and they follow up on them. So we send a letter to NHTSA explaining what we think they should do. Federal agencies are required to respond to us in 90 days. Nobody else has that kind of requirement, but we expect people to respond in a reasonable time span.

Then we continue to have dialogue with recommendation recipients periodically. We assume that some of the things we ask for are long-term goals, but we keep up with them: “How are you doing?” “What’s going on with this?” “When are you going to have a report?” Sometimes we’ll have a recommendation that we’ll consider an open acceptable response, which means we’ve discussed it with the recipient, they agree with the recommendation, they’re in the process, they’re doing the research, they’re making a design change — whatever is necessary. Sometimes it takes a long time for it to come to fruition, but at least the process has begun.

Have the recommendations from the March 2000 bus-train crash near Conasauga, Tenn., been heeded?
In that accident, there are 10 recommendations total. Seven of them are already closed as having been implemented. Three of them are still open, but those three actions are in fact being taken. A couple of them are being taken by NHTSA. They’re establishing performance standards on passenger protection for sidewalls and sidewall components and seats, so this isn’t something that occurs overnight. It’s a process. There’s a couple of recommendations like that to NHTSA that are “study and make changes.”

The other one that’s still open is one that went to all 50 states, and it’s a five-part recommendation. It involves installing stop signs at passive crossings; upgrading passive crossings with active warning devices; installing noise-reducing switches on new buses; enhancing bus driver training and evaluation; and including grade crossing questions on CDL exams. It’s a large recommendation; it went to all 50 states. I can tell you that, to date, 21 states have already made sufficient progress to justify closing this recommendation, and 15 more states are making significant progress. Thirteen other states and the District of Columbia have not made significant progress. One state, Hawaii, has no passive crossings. Some of the states’ efforts require legislation. So this is one of these long-term ones.

So there were 10 recommendations. Seven have already been implemented. Three are still open, but work is well on it’s way on all three.

What about the recommendations from the October 2001 school bus accident in Omaha, Neb.?
That’s the one where the bus fell off the bridge. There were seven recommendations, and they’ve all been accomplished. There were some school bus-related issues there, but there was also a number of highway work zone issues, and those have all been addressed. Not just locally, because we’re looking way beyond the local. This was a federally funded bridge construction project. There are national guidelines on how you set up work zones. This one met the letter of the work zone guidance but not the intent. So you have to look at it carefully, redefine some of your guidelines, and all of those things are occurring. And that involved federal agencies and associations, so it’s quite an effort. But all of these have been cleared up, and I think we have some impact on work zone safety throughout the nation from a school bus accident in Omaha.

What type of experience is needed to serve at the NTSB?
For investigators, we hire people with expertise in particular areas. If I need a mechanical engineer, I’m going to hire a mechanical engineer. I want him to have lots of knowledge. I want him to have some investigative expertise, and I want him to have been successful wherever he came from. We’re not opposed to hiring young people, but we often hire people who are 40 years old, because it’s taken them that long to have the depth and breadth of knowledge that makes them a cracker-jack investigator. Also, in the Office of Highway Safety, we have a lot of ex-police officers who have accident investigation experience.

Board members are people who are appointed by the president and confirmed by the Senate. There are five positions. Those people have some broad experience in the government and/or transportation safety. We look for senior people who have come from some other place and have knowledge of policies used in transportation safety.

Have you been directly involved in any school bus-related investigations?
Yes. As an investigator, I was the human factors investigator in the Fox River Grove accident. Over the past nine years or so, I’ve been involved in all of them either as the chief of investigations, in which I oversee all the investigations, or as the deputy director and now as the director.

What is your impression of the pupil transportation industry’s attention to safety?
That’s an easy one. It is second to none. Period. I deal in all sorts of transportation. The people in the school bus industry are not like any other industry. They are the most safety-oriented and energetic people that I know of. They take their charge very seriously, and it’s a pleasure to work with them.