On March 15, 2018, a partially constructed reinforced-concrete pedestrian bridge in Miami collapsed during a posttensioning operation onto Southwest Eighth Street, killing five vehicle occupants and one construction worker. The bridge was a planned connection between Florida International University’s Modesto A. Maidique campus in Miami and the city of Sweetwater. The collapse occurred 30 days after cracking was first reported on the structure and five days after the span was moved from the casting yard to what was to be its permanent location.
As a result of what a spokesperson has called safety failures related to the collapse, the Federal Highway Administration on July 14 of this year temporarily suspended FIGG Bridge Engineers of Tallahassee, Florida, the firm responsible for the bridge’s design, and William Denney Pate, P.E., M.ASCE, the senior FIGG bridge engineer who served as the engineer of record, from receiving contracts for or working on future federally funded projects. Additionally, the FHWA is considering debarring FIGG and Pate from working on federally funded projects for 10 years, a period more than three times longer than what is typically sought in such situations.
The suspension and potential debarment cover all federal contracts or subcontracts worth more than $35,000.
The FHWA’s suspension of FIGG and Pate took effect immediately. The debarment proposal must be decided on by Hari Kalla, P.E., M.ASCE, the FHWA’s suspension and debarment official and the associate administrator for infrastructure. At press time, Kalla had not yet made a decision about the debarment. An FHWA spokesperson told Civil Engineering that public safety is always the FHWA’s top priority, and the suspension and the debarment request are meant to prevent similar incidents from happening in the future.
Funding for the FIU pedestrian bridge came from multiple sources, including a Transportation Investment Generating Economic Recovery, or TIGER, discretionary grant from the U.S. Department of Transportation. Within the USDOT, the FHWA oversaw the FIU bridge project.
Immediately after the collapse, the National Transportation Safety Board — an independent federal agency dedicated to improving aviation, railroad, highway, marine, and pipeline safety — opened an investigation into the probable cause of the collapse. The NTSB’s final report was adopted on Oct. 22, 2019, and determined that the probable cause of the bridge collapse stemmed from design flaws. FIGG disputes this finding.
The FIU pedestrian bridge that collapsed was an innovative design that included, in part, the 174 ft long reinforced-concrete truss span that crossed Southwest Eighth Street. This span was formed from a canopy and deck walkway (comprising the truss’s top and bot
tom chords) connected by a single line of upright and irregularly shaped and angled diagonal truss members that extended down the centerline of the crossing. The upright and diagonal members were numbered 1 to 12, with the count beginning at the south pier and finishing on the north side. The connection points of two of these members with either the deck or the canopy were termed nodes.
The concrete deck was posttensioned in the transverse and longitudinal directions, and the canopy was posttensioned in the longitudinal direction. The diagonal and upright truss members were either permanently posttensioned (members 3, 5, 6, 7, 8, and 10), temporarily posttensioned for construction conditions (members 2 and 11), or not posttensioned (members 1, 4, 9, and 12).
The truss bridge segment was fabricated and constructed parallel to Southwest Eighth Street.Phased concrete casting was used, as first the deck, then the diagonal and upright members, and finally the canopy were placed. This resulted in cold joints at the top or bottom of each of the truss members’ nodes.
Both the Florida Department of Transportation’s Standard Specifications for Road and Bridge Construction and the American Association of State Highway and Transportation Officials’ AASHTO LRFD (Load and Resistance Factor Design) Bridge Design Specifications call for cold joints to be roughened, with the latter calling for that roughening to have a 0.25 in. amplitude. Both documents were referenced on the first page of the FIGG-prepared plans, and the plans specified this amplitude at the pylon diaphragm connections points. However, the plans did not specify this 0.25 in. amplitude surface roughening at the cold joints between the truss members and the deck or canopy. Though this detail was not contained in the plans, on June 13, 2017, Bolton Perez and Associates Consulting Engineers — the firm hired by FIU to administer, monitor, and inspect the bridge as it was constructed — sent an email to FIGG and the design-build contractor Munilla Construction Management Inc. about the cold joints’ construction. In its response, FIGG clarified that the cold joints needed to be roughened per FDOT specifications.
The concrete truss span was moved from its casting yard supports and into a simply supported position above the road on March 10. The move was attended by FIGG and was conducted using self-propelled modular transporters that lifted the span from its temporary shoring parallel to the road, rotated it, and moved it forward to lower it onto its permanent piers and final position across the roadway. It was the largest pedestrian bridge moved in this manner by SPMTs in U.S. history, according to the material circulated by the university before the collapse.
During the move, the bridge span exceeded the specified allowable torsional deformations twice. The sensor information was displayed in real time by a monitoring system so that corrective action could be taken as necessary, according to the NTSB final report. (Although FIGG staff members were on-site for the move, they were not notified of this exceedance until after the collapse). Within hours after the move, members 2 and 11 were de-tensioned, as planned, according to the NTSB final report.
From mid-February through mid-March, BPA sent three reports about cracking on the bridge to MCM. The first was sent on Feb. 13.
During the move, the bridge span exceeded the specified allowable torsional deformations twice.
After this first report, on Feb. 24, workers heard a loud noise, which the NTSB describes as the “distinct sound of concrete cracking” during formwork removal. Upon immediate inspection, a noticeable crack was found in the member 11/12 nodal region at the truss’s intersection with the deck. The second BPA crack report, on Feb. 28, mislabeled the general location of the cracking on the bridge, and FIGG notified MCM about this mislabeling on March 7.
In the days after the March 10 bridge move, the cracks significantly worsened. On March 13, BPA submitted to MCM its third crack report, an email it had sent on March 12 to FIGG, which included an image showing a 0.75 in. wide crack that was at least 3 in. deep at the northern end of the deck, according to the NTSB. This width is more than 40 times larger than the 0.016 in. width generally considered acceptable in reinforced-concrete structures, according to the NTSB’s final report.
Based on this emailed crack report and additional information gathered over the phone that the cracks had not increased in size after the span’s move, FIGG emailed instructions to MCM on the morning of March 13 to incrementally restress member 11 until it reached its original stressing force from the casting yard. In an email later that day, FIGG stated that it had “evaluated this further and confirm(ed) that this is not a safety issue.” Upon receipt of that email the next morning, March 14, MCM began coordinating the restressing of member 11 for the next day, March 15.
On the morning of March 15, the day of the collapse, Pate and another FIGG employee visited the main span at about 8 a.m. to view the cracks at the member 11/12 nodal region in person. This inspection was before a 9 a.m. meeting between FIGG, FDOT, FIU, MCM, and BPA to discuss the cracking. According to the NTSB final report, Pate presented at that 9 a.m. meeting that “there is no safety concern relative to the observed cracks and minor spalls” on the bridge.
The restressing of member 11 in its new position across the roadway, which was underway on March 15 when the span collapsed at about 1:46 p.m., was not shown on the design plans prepared by FIGG, according to the NTSB final report. Pate stated in a March 20 interview with the NTSB that the restressing was intended to bring the main span back to its “preexisting condition,” and as such was not independently peer-reviewed as a design change. The NTSB report disagreed with this statement because the main span’s supports in the casting yard were different than those used in the span’s position across the roadway and the cracking and dislocation of the upper portion of the nodal region had changed the geometry of the system. According to the NTSB, as a design change, the restressing should have been independently peer-reviewed per FDOT’s Structures Manual, Structures Design Guidelines.
Immediately after the collapse, FIGG hired Wiss, Janney, Elstner Associates Inc. to investigate the reason for the bridge’s failure with the understanding that the investigation would be led by Gary Klein, P.E., NAE, M.ASCE, executive vice president and senior principal at WJE.
The NTSB and WJE investigations differed significantly in scope and findings about the collapse. While both agree that the failure of the joint between the member 11/12 node and the bridge deck caused the span to collapse, the reason for that nodal failure is in question. On one side, the NTSB — with supporting work from the FHWA — looked at the entire project, identifying a number of issues in the bridge’s design, peer review, and handling of the structural cracking in the days leading up to the collapse. WJE looked specifically at the bridge move’s tilt exceedance, the collapse’s failure pattern, the peer-review process, and construction of the member 11/12 cold joint. The firm constructed and then tested full-size specimens of the member 11 connection to the deck in two conditions, one with and one without roughening of the cold joint. (WJE’s forensic investigation did not examine how the cracking was managed in the weeks, days, and hours leading up to the collapse.)
The NTSB concluded that a series of failures caused the bridge to collapse. Design errors in the bridge led to the separation at the cold joint at members 11/12 and the deck, and it could have separated even if the joint had been roughed to a 0.25 in. amplitude, according to the NTSB’s final report. The NTSB also noted that an inadequate peer review by global engineering and construction services firm Louis Berger missed this issue. In addition, FIGG’s mishandling of the cracking on the bridge as it formed and worsened also contributed to the collapse, the NTSB concluded. And the failure of MCM, FIGG, BPA, FIU, and FDOT to cease bridge work and close Southwest Eighth Street when the structural cracking reached unacceptable levels contributed to the severity of the collapse’s outcome.
Before issuing its final report in October 2019, the NTSB reviewed submissions from the FHWA and FIGG and discounted FIGG’s conclusions that its design was to code but improperly executed.
WJE’s forensic investigation concluded that the failure of the member 11/12 cold joint was one of construction rather than design. WJE says that its tests demonstrate that had the surface of the hardened concrete been roughened in accordance with FDOT’s Standard Specifications for Road and Bridge Construction, as required by the project construction specifications, the joint would not have failed.
Before issuing its final report in October 2019, the NTSB reviewed submissions from the FHWA and FIGG and discounted FIGG’s conclusions that its design was to code but improperly executed. After the NTSB issued its final report, FIGG contested the NTSB report’s findings in a heated written exchange with the FHWA that lasted through the end of January 2020, arguing that the WJE forensic findings were being ignored. (The FHWA suspension and debarment proposal of FIGG and Pate was issued about five and a half months after this exchange.)
On Aug. 11, FIGG and Pate filed, in federal court in Washington, D.C., a temporary restraining order against the FHWA and Kalla to lift the suspension. The filing argued that the FHWA’s suspension posed irreversible, “catastrophic and imminent harm” to FIGG and Pate.
That motion was denied in an Aug. 17 memorandum opinion by U.S. District Judge Colleen Kollar-Kotelly. She wrote in her decision that FIGG and Pate “have not demonstrated a likelihood of success on the merits, irreparable harm absent preliminary relief, or that the balance of the hardships and the public interest weigh in their favor.”
FIGG and Pate’s federal court filing argued that “the (FHWA’s) Suspensions (of FIGG and Pate) are not based on the Agency’s independent factfinding and fail to address in any meaningful way the competing evidence available to it on the face of the existing administrative record.” The FIGG and Pate filing continued, “Indeed, the record reflects a stubborn obstinance on the part of FHWA to even consider technical information presented by the Plaintiffs’ independent expert.”
In her rejection of the FIGG and Pate filing, Kollar-Kotelly pointed to the FHWA’s Jan. 23, 2020, letter to FIGG from Joseph Hartmann, Ph.D., P.E., M.ASCE, the FHWA’s director of the Office of Bridges and Structures. As quoted by the federal judge in her motion, Hartmann explicitly told FIGG that:
“I have found no new data or information in the material from FIGG/WJE that was not previously considered by the Federal Highway Administration (FHWA) in advising and supporting NTSB as it analyzed the facts and reached a determination of probable cause. … In this case, no uncertainty exists. The FHWA fully supports the findings and determinations of probable cause made by NTSB. The design errors by FIGG were the probable cause of the collapse of FIU pedestrian bridge, with the FIGG Engineer of Record contributing to the collapse by not appropriately recognizing the significance of the cracking that resulted from those errors.”
For More Details
• The National Transportation Safety Board’s final report, Pedestrian Bridge Collapse over SW 8th Street, Miami, Florida, March 15, 2018 (Oct. 22, 2019), can be downloaded at ntsb.gov/investigations/AccidentReports/Reports/HAR1902.pdf.
• The Wiss, Janney, Elstner Associates Inc. investigative report, FIU UniversityCity Prosperity Pedestrian Bridge Project: Research and Analysis Related to Collapse during Construction, Miami and Sweetwater, Florida (Sept. 18, 2019), is available as part of the FIGG party submission to the NTSB investigation. The WJE report begins on page 139 of the full submission, which can be downloaded at dms.ntsb.gov/public/62500-62999/62821/628567.pdf.
• FIGG Bridge Engineers, Inc. & William Denney Pate v. Federal Highway Administration and Hari Kalla filing (Aug. 11, 2020) can be downloaded at assets.documentcloud.org/documents/7034297/FIGG-and-Pate-v-FHWAmemo-in-Supprt-Motion-for.pdf.
• U.S. District Judge Colleen Kollar-Kotelly’s Aug. 17, 2020, memorandum opinion can be downloaded at cases.justia.com/federal/districtcourts/districtofcolumbia/dcdce/1:2020cv02188/220908/20/0.pdf?ts=1597742305.
This article first appeared in the October 2020 issue of Civil Engineering.