Skip Ribbon Commands
Skip to main content
Bookmark and Share this page


Keynote Remarks at the Chairman's Luncheon - Transportation Research Board, 90th Annual Meeting, Washington, DC
Deborah A. P. Hersman
Transportation Research Board, 90th Annual Meeting, Washington, DC

Thank you, Dr. Vest [President of the US National Academy of Engineering].

It is an honor to share the stage with Chairman Oberstar and Administrator Garvey this afternoon. You've already heard about their work that has made America's roads, rails and runways safer.

If it weren't for Chairman Oberstar I wouldn't be here today... literally.

Not only has Jim Oberstar been a personal mentor since I started my career as a congressional staffer, but when he was a Congressional staffer he helped craft the bill that created the NTSB. And he has always been a champion of the NTSB's mission.

Jane Garvey has been a role model for many women in transportation. Aside from her many professional accomplishments, I can personally attest that she is a real class act.

Last week Dr. Gridlock referred to this event as a Lollapalooza of transportation, but I've always felt that this is the transportation research Olympics. Bob Skinner, Suzanne Schneider, and the TRB staff have done an amazing job putting this conference together.

Thank you for inviting me to be your speaker today, because the truth is, when people meet me on the job they usually say, "I hope I don't see you again."

I try not to take it personally because usually when you see someone from the NTSB we're at the scene of an accident and something has gone terribly wrong.

Usually when you see the NTSB, it's already too late.

For those of you not familiar with the NTSB, we have investigated thousands of accidents and incidents and issued over 14,000 recommendations over our 40-year history. Today, I'm going to talk about four investigations to give you a taste of some of the cross-modal lessons learned, ranging from complex to relatively simple.

Because the TRB audience is so diverse, hopefully this speech will be like a cafeteria with something for everyone. For operators, there's the complex challenge of building a positive safety culture. For designers and builders, there's a simple lesson in keeping good records. And for regulators, I want to convey to you that the decisions you make regarding exemptions to regulations are just as important as the regulations themselves.

Investing in safety is not discretionary; like justice, safety deferred is safety denied.

No matter what your role, when you leave today I hope you take away a new perspective on safety culture, record-keeping, and aging infrastructure, so you will never be in the position of being "too late."

The first accident occurred in June 2009. I was in my office when I heard about an accident on Metro's Red Line. As we drove through rush hour traffic to the crash, none of us could understand how two trains could collide on a system designed to keep trains apart.

I recall very clearly the chaos of the accident site.

On that warm summer evening, emergency responders, press crews, residents and politicians came together to assist, report, understand and eventually mourn on that overpass above the train tracks.

There were 9 fatalities. It was the worst accident in WMATA's history.

What's not as well known is in the five years before the accident, a handful of smaller accidents preceded it, including one nearby at the Woodley Park station, and the NTSB had more business with WMATA than all other transit properties combined.

What's even less known is that in 2005, four years before, Metro experienced two near-miss events within minutes of each other under the river between Virginia and DC. They stemmed from the same type of component failure that caused the Fort Totten accident.

Metro's engineers responded to the 2005 incidents by devising an effective test to detect the failure.

But nobody was minding the store, and WMATA failed to ensure that the new procedures were adopted and understood by their maintenance crews, so the opportunity to prevent the next crash was lost.

What Metro needed in 2005 was a safety culture that ensured that the lessons learned were communicated to all parts of the organization.

When the collision occurred in 2009, it was too late.

The second accident I'd like to mention occurred last September.

As commuters were arriving home and families were sitting down to dinner in San Bruno, California, a 30 inch diameter natural gas pipeline exploded.

You probably all remember the dramatic TV footage – and perhaps thinking to yourself, could something like that happen in my neighborhood?

A 28-foot section of the pipeline blasted out of the ground and landed 100 feet away.

Over an hour passed before the valves were manually closed and the gas stopped flowing.

Scores of homes were destroyed, and 8 people were dead.

Our investigators were told that the pipe involved in the explosion was a seamless factory manufactured pipe.

But even a layperson could see the patchwork of welds marking the pipe.

This misinformation was not a minor record-keeping oversight.

In the years since the pipe was put into service, decisions regarding inspections, operating pressures, and risk management plans were all based on facts that were just plain wrong.

While the investigation is still ongoing, earlier this month the NTSB issued urgent recommendations calling for hydrostatic pressure testing to establish the safe maximum operating pressure of the pipeline.

Some of you may know that requirements for hydrostatic testing already exist – requirements established in 1970.

Ironically, the San Bruno pipeline was exempted because it was installed in 1956, so it was grandfathered from the testing requirements.

The operator has not been able to produce documentation on the origins of the pipe, the installation of the pipe, or the early inspection of the pipe.

But no one realized this until after the pipeline exploded. And then it was just too late.

Chairman Oberstar, I know there are many highway interests here today and all of them are familiar with the 2007 bridge collapse in Minneapolis.

I don't have to get into the details of our investigation of this accident to tell you that when that bridge collapsed, it was already too late.

The cause of the accident was gusset plates that were under-designed.

Since we're on the subject of record-keeping, I thought I would mention that neither the bridge owner nor designer had copies of the detailed gusset calculations.

Ultimately, we interviewed a retired engineer who had copies of some preliminary gusset design calculations -- in his basement – that helped us verify why those plates were under-designed.

He held onto those records the way we all keep files and mementos from our lives – things we just never quite get around to throwing away. Things we figure might be important one day.

And they were important. As San Bruno and Minneapolis demonstrate, record keeping is critical.

You may be tempted to think that the computer age has solved this problem that everything you do is saved and stored forever. But if you're like me, you know that regardless of whether your filing system is paper or electronic, it's the ability to retrieve what you've filed that's most important.

So unless you want NTSB investigators sifting through boxes in your basement after you're retired, keep flawless records and make sure that those who come after you can find them.

The final accident I will mention occurred in December 2005.

Shortly after take-off from Miami, the right wing literally separated from the fuselage of a Chalk's Mallard airplane.

Losing a wing may seem like a freak occurrence -- but trouble at Chalks was widely known.

The wing of the 58-year-old airplane had been weeping fuel for months.

FAA inspectors knew employees sealed the cracks in the wing but nobody addressed the underlying structural problem.

The airline had an anemic safety culture and its employees knew that shortcuts were being taken.

In fact, 3 of the 6 captains at Chalks left the company in the year before the accident due to their concerns about safety.

And if that wasn't enough, DOT had identified Chalks as a carrier in financial distress 18 months before.

But nobody connected the dots.

Nobody prevented an unairworthy plane from taking off.

And when the wing fell off in flight –it was too late for the 2 pilots and 18 passengers on board.

Ironically, the FAA has a program for aging aircraft that requires greater safety oversight. Believe it or not, the five-decade-old Mallard was exempted from the program.

If the San Bruno pipeline was a case of grandfathering, it's bordering on senility when an airplane is too old for the aging aircraft program!

Let's go back to the three lessons learned...

In Chalks, the pilots, the mechanics, the management and even the FAA inspectors all had a role in establishing a positive safety culture.

Like WMATA, it's not enough to do just what falls under your job description. You need to think about how the job you do relates to the jobs of others.

Similarly, if you're going to build a transportation project that's going to stand for decades, as in San Bruno and Minneapolis, the foundation of future operating decisions may well be based on the original documentation. Make sure your records stand the test of time.

All of the examples I've mentioned have something else in common: advancing age.

A 58-year-old plane. A 55-year-old pipe. A 40-year-old bridge. And the youngest, the Washington Metro system – though just 33 years old at the time – was running on a system without available replacement parts.

We believe – and Americans have the right to assume – that our infrastructure is built to last. And indeed it does last. The Transcontinental Railroad dates back to the Civil War, and the Brooklyn Bridge was built in 1883.

We definitely got the extended warranty on some of those landmark projects. Unfortunately at the rate we're going we really needed the lifetime guarantee.

Congress has passed over a dozen extensions for the FAA and at least half a dozen extensions for highway programs, primarily because they cannot agree on tough issues like financing.

The decisions made, or not made, in the coming months affect all of us in this room but, more importantly, determine the transportation realities for all Americans.

The fact is, the outlook for increased funding for infrastructure projects is grim, and safety programs are in jeopardy as we face reduced federal spending.

Right now, the real question for all of us is how hard these programs will be hit.

It's not my role as NTSB Chairman to weigh in on funding sources, but when it comes to investing in safety, we can pay now, or we can pay later.

In my role at the NTSB, I'm often put in the position of being a truth teller. So as an agent of reality it's my job to tell you that the concept of a lifecycle no longer exists.

Just because the train or plane you design is meant to last 30, 40 or 50 years doesn't mean it won't be around for 75.

The NTSB will always be there in an aftermath of an accident to figure out what happened. But that's only half the job. The other half is prevention.

When the Metro trains crashed, when the San Bruno pipeline burst, when the I-35 Bridge collapsed, and when the Chalks wing separated, it was already too late.

But it's not too late to stop future accidents.

We know that history repeats itself when lessons are not well-learned.

Let's build on the work of Administrator Garvey and Chairman Oberstar by creating a culture of safety, making sure that aging infrastructure is not exempted from safety requirements, and keeping records not just for your successor, but for your successor's successor.

Taxpayers don't have the option to buy a warranty on their local transportation project. This room holds the best insurance policy for our transportation system: YOU – the professionals that design, construct, maintain and regulate.

Your legacy can be a transportation system that keeps America running – and running safer – for generations to come.

Thank you.