FIU Pedestrian Bridge — A Miscalculated Node That Sheared Off Over Live Traffic

Summary

On 15 March 2018, at about 1:47 p.m., the partially constructed main span of the FIU-Sweetwater UniversityCity pedestrian bridge crossing the eight-lane SW 8th Street in Miami, Florida fell onto live traffic stopped at a red light, killing six people and injuring ten. The 174-foot concrete truss span dropped roughly fifteen feet onto the vehicles below. The National Transportation Safety Board (NTSB), the federal investigating body, identified the probable cause without hedging: load and capacity calculation errors made by the bridge designer, FIGG Bridge Engineers, in the design of the main span truss. The failure began at a single location — the nodal connection where diagonal truss member 11 and vertical member 12 met the bridge deck.

The mechanism was an under-designed connection, not an exotic one. The bridge was a concrete truss, and like any truss its loads concentrated at the nodes. At node 11/12 the steeply inclined diagonal pushed a large horizontal force into the deck, and that force had to be resisted by shear along a construction cold joint between the diagonal and the deck slab. FIGG underestimated the demand on that interface and overestimated its capacity to resist sliding. The NTSB found the actual demand on the node was nearly double the designer's calculated value, while the calculated shear resistance was too high. The connection was overloaded from the moment the span carried its own self-weight; it had no reserve at all.

The warning was visible and ignored. After the span was set in place on 10 March, severe cracks opened at the north end, precisely at node 11/12, and grew over the following days. Photographs were emailed to the engineer of record. On the morning of the collapse, the project team met and concluded the structure was not compromised and there were no safety concerns. The roadway was never closed. Hours later, a post-tensioning crew followed instructions to re-tension the rods inside diagonal member 11 — re-clamping the very joint that was already failing. The re-tensioning broke the last of the connection, the diagonal slid off the deck, and the span came down.

What makes the FIU collapse a permanent case file is that nothing about it was hidden. The error was a routine interface-shear calculation on a non-redundant structure. The cracks were documented, measured, and discussed. An independent peer review existed but missed the error. A road full of motorists sat beneath a connection that the designer's own arithmetic had under-built by a factor of two, and which was visibly tearing apart in the days before it fell.

Timeline

2015-2016

Design and procurement

Florida International University procures the UniversityCity Prosperity pedestrian bridge under a design-build contract, with FIGG Bridge Engineers as designer of record and Munilla Construction Management (MCM) as lead contractor. The chosen form is a single-line concrete truss spanning SW 8th Street, an unusual non-redundant configuration with no direct precedent.

2016

Independent peer review let cheaply

Louis Berger is engaged for the required independent design review. The contract is reduced from an initial quote near $110,000 to about $61,000 and is performed quickly; critical truss nodes, including 11/12, are not fully checked.

Late 2017-Feb 2018

Span fabricated beside the road

The 174-foot main span is cast on site adjacent to SW 8th Street using Accelerated Bridge Construction (ABC), to be moved into final position with minimal disruption to traffic.

10 March 2018

Span moved into place

Self-propelled modular transporters lift and carry the span across the road onto its supports. Within hours, cracking appears at the north (canopy) end at the 11/12 nodal region.

10-14 March 2018

Cracks grow and are documented

The cracks widen steadily over several days. Photographs are taken and emailed; the engineer of record reviews them and concludes there is no safety issue, only repairs to be made. The road remains open.

15 March 2018, 9:00 a.m.

Morning meeting downplays the cracks

FIGG presents the cracking to the project team. The meeting concludes the bridge's structural integrity is not compromised and there are no safety concerns. No decision is made to close the roadway beneath the span.

15 March 2018, late morning

Re-tensioning ordered

To close the cracks, a plan is set to restore tension in the post-tensioning rods inside diagonal member 11. A post-tensioning crew begins applying force to the member 11 rods.

15 March 2018, 1:47 p.m.

The node lets go

As member 11 is re-tensioned, the cold-joint connection at node 11/12 fails in shear. The diagonal slides off the deck, the truss loses its end support, and the span collapses onto vehicles stopped at the light below. Six people are killed, ten injured.

March 2018

NTSB launches investigation

The NTSB opens a major highway accident investigation, recovering the structure and reconstructing the load path and the failed node.

2019

OSHA cites design deficiencies

The Occupational Safety and Health Administration finds structural design deficiencies, concludes the designer failed to recognize that collapse was imminent, and notes the peer-review firm was not properly licensed for the work.

22 October 2019

NTSB final report

The NTSB adopts Highway Accident Report HAR-19/02. Probable cause: FIGG's load and capacity calculation errors in the main span truss, with the inadequate Louis Berger peer review and the failure to recognize the significance of the cracking as contributing factors.

The Single-Truss Bridge and Its Critical Node

The bridge was designed as a single line of concrete truss carrying a walkway, with a triangulated web of diagonals and verticals running along the centerline beneath a canopy. Unlike a conventional bridge with two parallel girders or trusses, a single-truss form has no second load path. If one node fails, there is no adjacent member to redistribute the force. The structure was non-redundant by configuration, which the Institution of Civil Engineers later flagged as an uncommon design whose risks demanded heightened scrutiny rather than less.

In any truss, force concentrates at the nodes where members meet. Node 11 and 12 sat at the north end of the span, where a steeply inclined diagonal (member 11) and a near-vertical member (12) framed into the deck. The diagonal carried a large compression force, and the horizontal component of that force had to be transferred into the deck and resisted by shear across the interface between the diagonal's base and the deck slab. That interface was a construction cold joint — concrete cast against previously hardened concrete — whose sliding resistance depended on the roughness of the joint and the reinforcing steel crossing it. This was the entire load path for the north end of the span. The deck end support, the diagonal, and the cold-joint shear were one chain, and node 11/12 was its weakest link by design.

How the Node Was Under-Built by Two

The governing failure was an interface-shear calculation that went wrong in both directions at once. FIGG underestimated the demand at node 11/12 and overestimated the capacity of the cold joint to resist it. The NTSB found the actual shear demand on the connection was nearly twice the value the designer had calculated, while the computed sliding-shear resistance of the joint was too high for the as-built condition. The joint surface had not been roughened or detailed to develop the resistance the calculation assumed, and the reinforcement crossing the interface was insufficient for the true force.

Mechanically, this meant the node was overloaded from the instant the span bore its own weight. There was no live-load margin to consume, no factor of safety waiting in reserve; the connection was below its demand at dead load alone. When the span was set on 10 March, the joint began to slip and the surrounding concrete cracked — the structure announcing, in widening fractures, that the load path at node 11/12 could not carry what the geometry was delivering to it. The cracks were not cosmetic surface crazing. They were the visible signature of a connection sliding apart.

The independent peer review that should have caught the error did not. Performed under a sharply reduced fee and on a compressed schedule, it did not fully check the critical nodes. The one safeguard designed to find a designer's arithmetic mistake was itself too thin to find it.

The Reckoning: Cracks Read as Cosmetic, a Joint Re-Clamped Under Traffic

The NTSB reconstructed the failure as a connection collapse driven by a design error and sealed by a fatal misreading of the warning. The cracks at node 11/12 were extensive and growing — among the most severe ever observed on a concrete bridge that had not yet failed. The engineer of record, employed by the designer rather than as an independent check, interpreted them as a serviceability problem to be repaired, not a sign of impending collapse. At the 9 a.m. meeting on 15 March, the team concluded there were no safety concerns and left the road open.

The final act was the re-tensioning. To pull the cracks closed, the team directed the post-tensioning crew to restore force in the rods inside diagonal member 11. Tensioning those rods drove the diagonal harder against the already-failing cold joint, applying the precise force the connection could not resist. The joint sheared through, the diagonal slid off the deck, the north end lost its support, and the span fell onto the cars below.

OSHA's separate investigation found structural design deficiencies and concluded the designer failed to recognize that collapse was imminent. No villain, no defective material, no freak load contributed. A node was under-designed by a factor of two, a thin peer review missed it, severe cracks were read as cosmetic, the road stayed open, and a joint that was already failing was clamped tighter while traffic waited beneath it.

Contributing Factors

01

The critical connection was the entire load path on a non-redundant structure

A single-line truss has no second load path; the node 11/12 cold joint was the only route the north-end force had to ground. A single connection failure was therefore a total-collapse event. Non-redundant forms remove the margin that lets a structure survive a local error, so every node in them must be designed and checked as if its failure ends the span — because it does.

02

The governing interface-shear demand was underestimated and the capacity overestimated

The defining error was a calculation wrong in both directions: demand nearly doubled and capacity inflated at the same joint. Compounding errors on opposite sides of a single inequality erase the entire factor of safety at once. The most dangerous calculation is the one that is optimistic about both the load and the strength of the same connection.

03

The cold-joint shear transfer was detailed for less resistance than the design assumed

The sliding resistance of the interface depended on joint roughness and crossing reinforcement that were not provided to develop the assumed capacity. A connection is only as strong as it is detailed and built, not as strong as the formula credits. Interface-shear and cold-joint details must be roughened, reinforced, and inspected to actually deliver the resistance the analysis claims.

04

The independent peer review was underfunded and skipped the critical nodes

The one safeguard meant to catch a designer's arithmetic mistake was performed under a slashed fee, on a compressed schedule, and did not fully check the governing nodes. A peer review priced and timed below the work it must do is a formality, not a check. Independent review of a non-redundant structure must be scoped, funded, and verified to examine the connections that can bring it down.

05

Severe, growing cracks at the failing node were read as cosmetic, and the road stayed open

The structure gave days of explicit warning at exactly the location that failed, and the warning was classified as a repair issue rather than imminent collapse. Cracking concentrated at a primary load-transfer node is a capacity warning until proven otherwise. When a structure cracks at its critical connection, the conservative response is to remove the load — close the road and stop work — before diagnosing the cause.

Aftermath

The FIU pedestrian bridge collapse killed six people and injured ten, most of them motorists with no connection to the project who were stopped at a red light beneath the span. The NTSB's final report, HAR-19/02, named FIGG Bridge Engineers' load and capacity calculation errors as the probable cause, with the inadequate Louis Berger peer review and the failure to recognize the significance of the cracking as contributing factors. OSHA cited structural design deficiencies and the designer's failure to recognize that collapse was imminent, and noted the peer-review firm lacked proper licensing for the work; FIGG was later debarred from federal-aid projects, and a Florida board found the engineer of record negligent. The case sharpened federal and state guidance on the independence and rigor of design peer review, the calculation and detailing of nodal interface-shear in concrete trusses, and — most pointedly — the duty to close a roadway and stop work when a structure shows severe distress at a primary connection. The Federal Highway Administration and AASHTO communities elevated scrutiny of non-redundant pedestrian-bridge designs and of the warning-sign protocols that govern when a partially built structure must be unloaded. The collapse became the modern byword for a connection under-designed by arithmetic, a thin review that missed it, and a crack pattern that was photographed, measured, discussed, and walked past while traffic kept flowing underneath.

Lessons

Treat every node on a non-redundant structure as the whole load path: design and check the governing connection as if its failure ends the span, because without a second path it does.
Be most suspicious of the calculation that is optimistic about both the demand and the capacity of the same connection, since a paired error erases the entire factor of safety at once.
Detail and inspect interface-shear and cold-joint connections to deliver the assumed resistance — a joint is only as strong as it is roughened, reinforced, and built, never as strong as the formula credits.
Scope, fund, and verify independent peer review to actually examine the connections that can collapse the structure; a review priced and timed below the work is a formality, not a safeguard.
When a structure cracks severely at a primary load-transfer node, remove the load first — close the road, stop the work, evacuate the area — and diagnose the cause only after the danger is gone.