A Prefabricated Tragedy

The collapse of a superbridge in Florida shows how an entire philosophy of building can go wrong.

Law enforcement and members of the National Transportation Safety Board investigate the scene where a pedestrian bridge collapsed a few days in Miami, Florida.
Law enforcement and members of the National Transportation Safety Board investigate the scene where a pedestrian bridge collapsed in Miami.
Joe Raedle/Getty Images

It was to be a showcase of advanced bridge technology, a centerpiece of urban development in Miami-Dade County. It was equipped with titanium dioxide–impregnated self-cleaning concrete that would always sparkle white in the Florida sun. It was designed to withstand a Category 5 hurricane. And it was supposed to last 100 years.

But last week, the FIU-Sweetwater UniversityCity Bridge failed catastrophically during construction, crushing rows of cars stopped at a red light on a busy thoroughfare. Six people were killed and 10 injured.

Failures like this invoke special horror because they are so rare, which is the only solace to be had from the tragedy. As pedestrians and motorists, we take infrastructure reliability for granted in a way that we don’t with our consumer goods. We expect cars and computers to crash, and conceptually, we can cope with that, even when the consequences are dire. When bridges and tunnels implode, though, it seems like things are truly out of our hands, the sky almost literally falling on our heads.

So it is a tad disconcerting to learn that the technology and techniques used in the UniversityCity Bridge may be coming to a neighborhood near you, if they haven’t already. The bridge was a product of something called “accelerated bridge construction,” or ABC, a technique of fast-tracked prefabricated building that has strong political backing at both the state and federal levels. More than 1,000 bridges have been built with it, and Florida International University is one of the leading research centers in this kind of engineering, an irony not lost on some commentators.

To be fair, the university did not design or build the bridge, as it hastened to point out. And neither the designer nor the builder has a spotless record. With wreckage left to be cleared, it will be a while before the National Transportation Safety Board can pinpoint the exact cause of the disaster.

What we can say is that when complex technological systems fail, usually no single factor is to blame. In her study of the Challenger Space Shuttle disaster, the sociologist Diane Vaughan noted a phenomenon she called the “normalization of deviance,” an acceptance of assumptions and shortcuts that over time incrementally piles on risk until, like compounding interest on debt, a kind of technological bill suddenly becomes due.

Something like this probably happened at the UniversityCity Bridge. And when all the rubble is removed and the reports are written, it is likely we’ll learn that the collapse had something to do with the philosophy of accelerated construction. In a part of the country notorious for slash-and-burn urban development, it turns out that ABC can be much more complicated and problematic than its rather glib acronym implies.

First, let’s discuss what probably didn’t cause the collapse. The FIU bridge looked like a cable-stayed bridge, a highly reliable technology that has been around since at least the 16th century. Cable-stayed bridges look something like suspension bridges, but the physics are significantly different. In suspension bridges, the load is transmitted through two sets of cables, one set attached to the deck that is in turn connected to two main cables loosely strung between towers, and passes into the ground at the ends, where the cables are anchored stoutly. As anyone who has walked or driven over one can attest, suspension bridges are relatively flexible. They move, and even sway, in heavy traffic, strong winds, or earthquakes.

In a cable-stayed bridge, everything is a lot stiffer. There is only one set of cables and they are rigidly embedded into towers and the deck, which has to be a lot more robust than a suspension-bridge deck because the wires tend to exert strong horizontal forces. This design is favored when ground conditions aren’t suitable for the kind of massive anchors suspension bridges require. Such bridges require fewer materials and can be erected more quickly than their suspension counterparts.

Cable-stayed bridges are also more flexible than suspension bridges in terms of design and can be built in many different configurations that play on the arrangement of cables and towers. Boosters love them for their aesthetic potential, and some of the most prestigious recent bridge projects are of the cable-stayed variety, including Santiago Calatrava’s Margaret Hill Hunt Bridge and Sir Norman Foster’s Millau Viaduct. Cable-stayed showcase bridges have become increasingly popular in the U.S. and especially Florida, where they are regarded as expressions of civic upward mobility that no self-respecting aspiring metropolis wants to be without.

In the cable-stayed bridge, the tower is the thing. It bears all of the load, which it transmits to the ground, and it is also the dramatic fulcrum of the design. Suitably enough, in boat-mad Florida, the tower in the UniversityCity design was supposed to evoke a sailboat.

But the UniversityCity design was not actually a cable-stayed bridge. It only looked like one. It was actually a kind of truss, also a well-proven technology. The proposal diagrams suggest that the two primary deck sections could be assembled prior to the erection of the tower and cables. The main section weighed 950 tons, and was exceptionally large and heavy for a bridge intended to accommodate only pedestrians. It was swung into place on support pylons in a much-publicized operation employing a huge mobile jack. And there it sat for five days until it failed. Two days before the collapse, as the New York Times reported, small cracks were observed at the slab’s north end, where the tower was to have been installed and where the failure appears to have occurred. The cracks caused sufficient disquiet that engineers were discussing the structural integrity of the span, proclaiming it safe only hours before the collapse.

Understanding why the bridge was built and assembled in this way is a lesson both in the politics of prefab and civic boosterism. All first-year civil engineering students learn that concrete is good in compression and weak in tension. Steel is good in both. But concrete can be given tension by prestressing it—that is, by running steel cables through it. This can be done at the time concrete is cast, in what is known as pre-tension.

It can also be done after casting. This is known as post-tensioning and involves tightening and locking the cables that bind a cured slab. The UniversityCity Bridge used post-tensioned parts. In an ideal world, cable tensioning is part of the prefab process. For bridges that will bear heavy traffic, engineers want to achieve a slight upward bend called a camber, which flattens out under load and gives high carrying capacity. But sometime cables are overtightened, and engineers would want to resolve adjustments before final assembly. It seems that at the UniversityCity Bridge, the slab was stress-tested not before but after it was mounted in place. In a partly assembled structure, such a procedure is risky and has been known to end in catastrophic failure. Reports suggest that the collapse occurred when workers were adjusting the cables and, presumably, the camber.

Prefab assumes that all the complexities of construction can be front-loaded into component manufacturing, so that final assembly can take place all at once, quickly and efficiently. It has been marketed as the scientific management of building, a triumph of logistics over craft knowledge. Naturally, unions tend to hate prefab, seeing componentization as a means of destroying their ability to control the pace of construction. Managers and boosters love it for precisely those reasons.

At UniversityCity Bridge, it looks as if one of the basic precepts of prefabricated construction was undermined. If accelerated construction were premised as means of simplifying project management, prefab was surely never intended to be a complete substitute for it, as those in charge seem to have assumed. It looks as if a giant concrete Lego set was misassembled, with fatal results.

So what were the problems for which this application of ABC was the solution? Ultimately, they issue from the rush to develop some of the world’s most valuable but ecologically fragile real estate. To the pioneering activist Marjory Stoneman Douglas, the story of the colonization of Florida was one of greed, violence, displacement, and destruction on an epic scale, immortalized in her magisterial Everglades: River of Grass, a classic of environmental history.

Modern boosters, including FIU’s ABC center, tend to reduce Florida’s problems to traffic throughput. Western Miami-Dade was hacked out of the Everglades, which in the wet season boils over with freshwater that is kept at bay only with extensive public works. Foremost among these is the Tamiami Trail, the artery that the UniversityCity Bridge was to have spanned. Built in the 1920s to connect Florida’s east and west coasts, the road is basically a long, low dam with a highway on top of it. It cut off the flow of freshwater from Lake Okeechobee to the southern Everglades, drying up the wetlands and creating an environmental catastrophe.

Today, critics regard Tamiami Trail as emblematic of Florida’s shortsighted approach to urban development, although the U.S. Army Corps of Engineers has done remedial work to remedy the water flow problem. Yet it remains an important east-west highway in the southern part of the state, and so it is a strategic traffic resource. For visitors driving in from Naples, the Tamiami is Florida’s Appian Way, and the UniversityCity Bridge was to be its triumphal arch, a gateway to greater Miami. Construction couldn’t be allowed to block traffic, which is political plutonium in any auto-centric society, especially South Florida. Enter accelerated construction.

The stakes here are far larger than the ambitions of one Sun Belt city and university. The federal government has helped fund projects like the UniversityCity Bridge through its Transportation Investment Generating Economic Recovery initiative, a program that has drawn criticism for privileging politics over technical merit. And federal money has played perhaps the major role in terraforming the state, not least through the National Flood Insurance Program, 35 percent of which is devoted to the Sunshine State, the largest single chunk of the program.

So it is hard not to see the tragedy of the UniversityCity Bridge as a kind of metonym for the reckoning that Florida has long been setting itself up for. Designed to withstand the biggest storms nature could throw at it, the structure couldn’t withstand a perfect storm of hasty planning, managerial incompetence, and human hubris.

Correction, March 20, 2018: This article originally identified the FIU-Sweetwater UniversityCity Bridge as a cable-stayed bridge; it was a truss bridge. The article also said steel is weak in compression and good in tension; in fact, it is good in both. It described camber as a dangerous bend in the slab; in fact, it is a slight upward bend that is not necessarily dangerous. And it incorrectly described Tamiami Trail as the only east-west highway in South Florida.