Creeping Disaster: The Big Dig Tunnel Tragedy One Year Later

Just over a year ago, a woman died in the collapse of a part of the ceiling over a Boston highway tunnel that forms part of the so-called Big Dig. Less than a week after the collapse, experts were talking about how the epoxy used to hold up the ceiling tiles could fail. In the year that has passed since then, the National Transportation Safety Board investigated the accident and released their report on July 11, the one-year anniversary of the collapse.

At the time, I remarked on the apparent similarities between the Boston tunnel collapse of 2006 and the Kansas City hotel walkway disaster of 1981, in which 114 people died. As it turns out, the comparison was apt. In Kansas City, a contractor made an apparently innocuous change in the way some threaded support rods were arranged. But the change greatly weakened the structure and contributed directly to the collapse. The NTSB report says that while epoxy can be used safely to hold bolts in place to support suspended ceilings in tunnels, the wrong kind of epoxy was used in the ceiling that failed.

Epoxy adhesives have been available in some form since the 1940s, but to recommend their use in critical structural elements such as multi-ton ceiling slabs, the manufacturer needs to understand short-term and long-term chemical and physical processes in the material. It turns out that in common with many other plastics, certain kinds of epoxy (including what the NTSB called "fast-set" adhesive) slowly stretches under stress. This behavior is called "creep," and my blog of July 19, 2006 noted that engineering experts were already speculating that creep might have been responsible for the collapse.

It was. The epoxy vendor Powers Fasteners also sold another kind of epoxy, "standard-set," to the Big Dig contractor, Modern Continental Construction Company, intending it to be used for the critical ceiling bolts. Unlike the fast-set type, the standard-set epoxy does not creep when installed properly. Bechtel/Parsons Brinckerhoff, the consulting firm overseeing specifications for the project, allowed Gannett Fleming Inc., the ceiling designer, to specify the adhesive by performance rather than a particular make from a particular company. Such a practice is in keeping with the competitive-bid process, but often makes it harder to tell what is really needed for a specific job.

Of all the entities involved—the designer, the contractor, the vendor, and the people on the ground who actually put the adhesive in the holes—the NTSB found that only the vendor, Powers Fasteners, understood the danger of creep and the need to use the non-creeping standard-set epoxy, not the creep-prone fast-set type. But somewhere along the line, possibly under deadline pressure, that vital bit of information got buried in fine print, someone substituted the fast-set epoxy, and the deadly chain of events was set in motion.

If the Massachusetts Turnpike Authority, the organization responsible for operating the tunnel, had carried out prompt and thorough inspections of the tunnel after it opened, they would almost certainly have discovered signs that the bolts were creeping out, and could have taken corrective action. But the NTSB found that before such regular inspections could take place, the MTA felt obliged to compile a database of tunnel components and apply to the Federal Highway Administration for approval of its inspection plan before putting it into action. This bureaucratic musical-chairs performance took three and a half years—longer than the ceiling took to creep out and collapse.

There are many ironies in this episode, but I will content myself with pointing out two.

First, right in the heart of what in my less charitable moments I refer to as the "know-it-all capital of the world," the land of Harvard, MIT, and one of the greatest concentrations of engineering experts in the world, a critical life-saving bit of knowledge—the information about creep—didn't get to the people who were in a position to do something about it. I teach at an institution that is to Harvard or MIT as a culvert under a farm road is to the Big Dig. But we have a large construction program here, where hundreds of students learn the basics of materials and other dry matters on their way to becoming foremen and supervisors of the same kinds of workers who put the wrong epoxy in the ceiling in Boston. I can only hope that if our students were in the same position, they would have known better. I dare say MIT, or even Massachusetts as a whole, does not pay much attention to students who want to be contractors when they graduate. But if humble construction education programs such as ours teach people in that line of work about the dangers of ignorance when it comes to novel materials, we will have justified our existence in that regard, anyway.

Second, the kind of bureaucrat who values procedure and compliance and following all the rules above simply doing the right thing, is not serving anyone well in the long run. If there had been just one low-level inspector or employee of the MTA who had said to himself, "The hell with waiting forever for FHA approval—I'm going out there and take a look," he might have found the problem early enough to forestall it. But he would have had to raise a big stink, probably go over the heads of his supervisors, perhaps even go to the media, and in all likelihood he would have lost his job. Such people are called whistleblowers, and they are the engineering world's equivalent of the Old Testament prophet—one who speaks the truth regardless of how unpopular it might be, or how dangerous it is to one's own well-being. Like the office of prophet, it is a lonely calling, one that should not be entered into lightly. But paying the price of unpopularity, or even sacrificing one's career, is small compared to the saving of lives.

Sources: Articles describing the NTSB report were carried by the Boston Globe on July 11, 2007 (http://www.boston.com/news/local/massachusetts/articles/2007/07/11/wide_risk_wide_blame/) and the New York Times (http://www.nytimes.com/2007/07/11/us/11bigdig.html?_r=1&oref=slogin).