On March 15, 2018, a 175-foot-long (53 m) section of the FIU-Sweetwater UniversityCity Pedestrian Bridge collapsed while under construction. The collapse resulted in six deaths (one worker and five motorists), ten injuries (six serious and four minor), and eight vehicles being crushed. One employee was permanently disabled. At the time of the collapse, six lanes of road beneath the bridge were open to traffic.
The pedestrian bridge was designed to connect the town of Sweetwater to the campus of Florida International University (FIU) in University Park, a suburb west of Miami, Florida, United States. The two were separated by a busy eight-lane highway, which the bridge was designed to span.
The engineering design error that directly led to the collapse was identified by the National Transportation Safety Board (NTSB) as a miscalculation of resistance to sliding of the connection between the walkway surface and the truss that held it up. The walkway surface was poured concrete, which was allowed to harden before the truss braces were poured above it. These truss members were connected to the deck by steel reinforcing rods embedded in the deck and in the concrete of the truss. To hold up the bridge, these connections had to prevent the truss from sliding along the walkway surface. The resistance to sliding was miscalculated, and was insufficient to prevent the connection from sliding and causing cracks in the truss concrete. Cracking ultimately caused the complete disconnection of one of the truss-to-walkway connections, leading to the collapse.
Background
The FIU Sweetwater UniversityCity pedestrian bridge, located just west of the intersection of Tamiami Trail (Southwest 8th Street) and Southwest 109th Avenue, was planned to connect the FIU campus to student housing neighborhoods in Sweetwater. It was intended to improve pedestrian safety, as the crosswalks at this wide, busy intersection had been identified as a safety hazard. One student had already been struck and killed by a vehicle there. The $14.2 million project was funded with a $19.4 million Transportation Investment Generating Economic Recovery (TIGER) grant from the United States Department of Transportation in 2013, along with funding from state agencies. The bridge cost $9 million to construct, exclusive of the installation cost.
The main companies behind the construction project were Munilla Construction Management (MCM), a Miami-based construction management firm, and FIGG Bridge Engineers, a Tallahassee-based firm. Unlike most bridges in Florida, the design for this project was overseen by the university, not the Florida Department of Transportation (FDOT), in a program known as the Local Agency Program (LAP).
Florida International University is known for its expertise in accelerated bridge construction (ABC) and has attracted international scholars as PhD students. It is home to the federally funded Accelerated Bridge Construction University Transportation Center, which sponsors industry conferences and seminars. The National Transportation Safety Board, however, found that with respect to the bridge, "F.I.U. had no professional engineers on its staff and relied solely on the expertise of its hired contractors."
Bridge layout
The full 320-foot-long (98 m) pedestrian bridge was to cross both a major roadway and a parallel water canal with two separate spans connected at a faux cable-stayed tower. The main roadway-crossing span was 175 feet (53 m) long, and the shorter canal span was to be 99 feet (30 m) long. An elevator and stairs at the south end added 31 feet (9 m), and at the north end, 15 feet (5 m), for a total bridge length of 320 feet. At the bridge site, the Tamiami Trail roadway had seven lanes of traffic and one turn lane.
Bridge design and construction
The new pedestrian bridge was designed to connect the campus to student housing in a dramatic, sculptural way and also to showcase the school's leadership in the ABC method of rapid bridge construction. The bridge was meant to last more than 100 years and to withstand a Category 5 hurricane, according to a statement by the university.
The full bridge project was styled to look like a cable-stayed bridge, with a pylon tower and high cables for dramatic effect. Functionally and structurally, it was a mono truss bridge, with the spans being fully self-supporting. The bridge spans used a "re-invented I-beam concept", a novel concrete truss design invented for this project. Concrete truss bridges are "exceedingly rare," and as of 2018, "no other designs similar to the FIU bridge" had been discovered. The vertical web of the beam consisted of a series of triangulated concrete diagonal struts along the centerline. The diagonal angles of the struts varied across the bridge so they would align with pipes from the center pylon in the eventual faux cable-stayed appearance. The concrete walkway deck was to act as the horizontal bottom flange of a wide I-beam, and the concrete roof canopy was to function as the horizontal top flange of the I-beam. The walkway was thus nearer to ground level than in a standard design where the walkway is placed on top of the structural support system. This reduced the number of steps that bridge users would have to climb.
The bridge was a post-tensioned concrete structure. Concrete structures are generally ten times heavier than equivalent steel designs. The bridge was made using a new formulation for concrete, which was intended to stay cleaner than standard concrete formulations. In the main bridge span, the concrete floor deck, roof, and most diagonal struts contained post-tensioning (PT) members whose compressive effect on the concrete was adjusted after the concrete was cured but prior to loading.
The deck and the canopy had PT cables that were permanently tensioned after the concrete was cured and while the main span was still located in the staging yard. Most of the diagonal truss members, also known as tendons, had two to four PT rods that could be tightened or loosened depending on the engineer's design plan. The high-strength steel rods are sleeved to allow free movement, with only their ends in contact with the concrete to provide a clamping force. This applied tension is necessary to keep concrete members in compression. All of the adjustable tendon rods were oriented with the nut end on top of the canopy, hidden with a "blister" of extra concrete. Adjusting the rods generally required a crane and the specialized crew of a subcontractor.
Two tendons of the main span, particularly the ones nearest the support ends, had PT rods designed for temporarily prestressing the concrete tendons during the movement of the 950-ton span. These four PT rods were tightened explicitly for the day it took to transport the main span over 8th Street and de-stressed (loosened) immediately thereafter. The purpose of the temporary prestressing of the two end tendons was due to the different loads during movement. The transporter's four-point support of the massive span essentially forces the two terminal ends of the bridge span to be cantilevered past the supports, unlike its final placement. To keep the bridge stable during transport, the four PT rods were torqued to each apply a load of 280,000 lbf (1,245,500 N). The other PT rods, inside the other tendons, were tightened just once and considered permanent. The transport of the prefabricated, 950-ton span (see below) to its location over 8th Street was performed five days before the subsequent collapse.
A specification change from the FDOT late in the planning phase required relocating central pier 11 feet (3.4 meters) north to allow for future widening of the highway, causing some changes in construction.
Construction of the bridge began in March 2016 and was scheduled to be completed in December 2018. The bridge's main span was assembled adjacent to the highway using accelerated bridge construction. It was lifted into place on the morning of March 10, 2018, five days before the collapse, during a weekend closure of the roadway.
Collapse
Reports of cracking
The main span of the bridge was in place by around 11:30 am on March 10. Inspections were performed, and the order was given to release the tension on the steel post-tensioned (PT) tendons at the terminal ends of the span. Kevin Hanson, the supervisor of the de-stressing operation of the PT tendons, became aware that cracks formed on the north side of the bridge and was concerned. He sent a text message to his supervisor, Sam Nunez, stating that "it cracked like hell" and included photos of the location near where the detensioned tendons were located. Pedro Cortes and other members of the construction contractor reviewed the cracking and took photos. These cracks were located in the area where the bridge contacts its supporting pier, where the structural diaphragm deck was located. Other cracks were observed in the northernmost diagonal truss member.
Two days before the collapse, on March 13, the engineer of record (EOR) from FIGG became aware of the cracks and began calculating remedial action, including the addition of plastic shims under the diaphragm. He then reported the cracking by voicemail to an FDOT employee. He thought this was not an immediate safety issue, merely something that would need to be repaired later. The FDOT recipient was away for several days and did not hear this message until the day after the collapse. FIGG's engineer of record then decided to re-tension the temporary PT rods in the northernmost tendon to their state on March 10. This plan was developed by FIGG over the 13th and 14th from the FIGG office in Tallahassee, and a meeting was scheduled for March 15 at 9:00 am in a trailer at the construction site.
Two FIGG engineers, including the EOR, arrived before the meeting to inspect the bridge span along with some of the managers of the consulting engineering firm and the main construction company. They used a lift to get a close view of the damage.
At 9 a.m. on March 15, a university employee heard a loud "whip cracking" sound while under the bridge span, waiting for a red traffic light. At the same time, the design-build team met for about two hours at the construction site to discuss the cracks discovered on March 10. Representatives from both FIU and FDOT were present. The FIGG lead engineer's conclusions were that the structural integrity of the bridge was not compromised and that there were no safety concerns raised by the presence of the cracks. FIGG also insisted that no crack repairs should be carried out until the stabilizing of the node and pylon diaphragm with post-tensioning was completed.
An additional measure proposed by FIGG at the meeting to remediate the cracking was telling the contractor that it "must expedite the pouring of the intermediate pylon," a structure that was designed to mimic the tower of a cable-stayed bridge. This pylon was to be located on the north end of the main span adjacent to the area that failed and affixed to the same substructure that the failing main span diaphragm sat upon.
The FIGG engineers on site "did not know the reason for the cracks, but still expressed no safety concerns" at the meeting.