On 15 March 2018, a pedestrian bridge connecting Florida International University (FIU) to the nearby town of Sweetwater collapsed on top of traffic across a 6 lane roadway below. Six people lost their lives and ten others were sent to the hospital. The bridge, which was installed just 5 days prior to the collapse, was designed to withstand a category 5 hurricane. The 174 foot span concrete truss structure weighing 950 tons was built using the Accelerated Bridge Construction (ABC) method. After the event, questions were publicly raised asking whether the ABC method may have contributed to the failure, but a resilience engineering approach must also consider how issues like decompensation due to cascading events and conflicting drivers like schedule, budget, and design changes are involved.
ABC Bridge Design Method
One reason the ABC method is in question is because it is seen as a new construction practice in the US. Although ABC was only adopted in the US in the mid 2000s, it has been used in Europe for several decades. The Federal Highway Administration, promotes the increased use of ABC because the method aims to reduce onsite construction time, improve constructability, and save on costs. Benefits of the ABC method include:
- Faster delivery time—many parts of the project can be concurrently fabricated and shipped when ready for installation thereby reducing onsite construction time.
- Better site constructability—work can be done onsite parallel to existing structures while many of the bridge components can be assembled off-site in a climate controlled environment. This also improves work-zone safety for workers and travelers alike.
- Cheaper to build—although direct project costs may be larger, the total costs are considered cheaper because reduced on-site construction time means fewer traffic disruptions. Fewer interruptions to commercial and industrial activity minimizes the negative impact on local and national economies.
The Federal Highway Administration (FHA), also states that over 800 ABC projects received federally funding between 2010 and 2012. This number is small considering that about one fourth of the nation’s 600,000 (~150,000) bridges are in need of repair or replacement. This may explain why many consider the ABC approach as a newer technology.
Was the design method to blame?
At first glance, it seems reasonable to question the viability of the ABC method in relation to the bridge collapse. After all, the proposed benefits of faster, better, cheaper represent a conundrum in the resilience engineering of complex adaptive systems. Numerous case studies have shown how a faster, better, cheaper strategy can compromise system performance and degrade safety margins. However, while the cause will remain unknown until after the NTSB investigation is complete in early 2019, there are some interesting considerations that suggest the unfortunate accident coincides with a sequence faster, better, and cheaper operational events that are unrelated to the design methodology.
These are ordinary events that could happen in any type of construction project. When viewing the design and construction process as a system, the failure occurred amid mounting empirical evidence of increased brittleness. In other words, numerous opportunities to take corrective action were missed. The inability to deploy adaptive capacities in a timely manner led to system level decompensation and tragedy.
Sequence of events leading up to the bridge collapse
Consider the following as reported by the Associated Press and other news agencies:
- Florida Department of Transportation intervened in the construction process by placing new requirements to relocate one of the bridge’s main support structures. FIU and the construction contractors were ordered to move the support structure 11 feet north toward the edge of the nearby canal to allow for future growth of the roadway. This also led to the removal of a temporary support beam.
- Moving the support structure widened the gap between the bridges main end supports, which required new design work, more time, and more money. The bridge was already behind schedule and millions of dollars over budget. The redesign effort exasperated these issues and created even more pressure to complete the construction as quickly as possible.
- Cracking appeared on the bridge near where the support structure and a temporary support beam were moved. The cause of the cracks and the role that they may have played in the failure has not yet been determined.
- The cracks were reported to the FDT by a lead engineer just two days prior to the collapse. The FDT official was out of the office at the time and the message was left on voice mail. The transcripts of the voice mail have been released under the freedom of information act revealed that the lead engineer was of the opinion that the cracks posed no threat to safety.
- Construction engineers met with a consultant representing FDT on the Thursday morning just hours before the collapse. At that meeting, the engineers gave a technical presentation on the cracks and stated that they did not believe the cracks posed any safety concerns or compromise to the structural integrity of the bridge.
- Around the time of the meeting, a university employee was passing near the bridge and heard a large cracking whip kind of sound. Later he would say that the look on the face of the nearby construction worker indicated that that was not a good sign. There is no information about what action may or may not have been taken as a result of the odd sound.
- At the time of the collapse, engineers were tightening cables on the North end of the bridge.
- Finally, the collapse started near the North end where the cracks were located and where the cables were being tightened.
Dashcam video of the bridge collapse
What to make of it all?
Why did this new bridge that was designed to withstand a Cat-5 hurricane and hailed as a breakthrough in fast, safe, and cost effective bridge construction collapse on a calm sunny afternoon five days after it was installed? There are no simple answers and I am certainly in no position to pass judgment. That is the job of the NTSB, project engineers, and other officials investigating the accident. However, it is interesting to note how the events leading up to the collapse are clearly set apart from the ABC method. The same sequence of events could have taken place in a regular approach to bridge design and construction.
From a resilience engineering perspective decompensation occurred when the breakdowns began to cascade faster than the engineers were able to respond. The whole sequence of events led to a tipping point whereby the well-intended adaptive capacities were grossly insufficient.
The breakdowns in communication and decision making led to missed opportunities that compromised safety margins. These events suggest that faster, better, cheaper strategies among the people who install, permit, inspect, or regulate construction in Florida may be the pressures that led to the collapse, not the ABC method itself. If this is the case, then the bridge collapse may be symptomatic of more sinister issues that plague the Florida construction industry, and more failed construction projects may soon result.
It will be interesting to see how the NTSB investigation considers operational deficiencies in contrast to the ABC design method and how the NTSB recommends improvements to safety practices. Although the ABC method may hold great promise on other future projects, a faster, better, cheaper strategy unrelated to construction practice can still compromise the resilience of a system.