A collapsed or cracked bridge span concentrates decades of traffic cycles, weather, and deferred maintenance into one failure. Determining which margin ran out — and who let it — is forensic engineering under real stakes.
Start a conversation with our AI Research Concierge, already scoped to bridge failure. Pick a starting point, or describe your situation directly.
Bridges are engineered with an expectation that any single member can lose capacity without the structure failing — until the systems that provide that redundancy are themselves compromised. A fatigue crack in a welded detail grows silently for years before it reaches critical length; a flood scours streambed material from around a pier footing in a matter of hours; a frozen bearing quietly forces a girder to resist forces it was never designed to carry. None of these mechanisms are exotic, and none of them are invisible to the right examination — but the evidence for each is different, and reading it correctly is what separates a defensible cause from a guess made from photographs.
Bridge failures are rarely one thing — they are a mechanism that closed a margin of safety accumulated, or lost, over years. Identifying which one applies determines everything that follows.
Millions of live-load cycles concentrating stress at welded and bolted details until a crack initiates and propagates, often at a detail with an inherently low fatigue category that was never flagged for monitoring.
Deicing-salt chlorides and moisture attacking exposed steel and reinforcement, thinning load-carrying sections gradually until the remaining area can no longer carry the rated load.
Flood flows removing streambed material from around piers and abutments, reducing the depth of embedment footings depend on until bearing capacity or lateral stability is lost.
Overweight permit loads, illegal loads, or an over-height vehicle or vessel strike delivering forces the member, or the whole span, was never rated to withstand.
Frozen, corroded, or improperly installed bearings restraining thermal movement and forcing unintended stresses into girders and substructure never detailed to resist them.
Underrated capacity, an outdated load rating that never accounted for a permitted route, or a detail that does not match the as-built condition it was assumed to represent.
Bridge investigations combine field non-destructive evaluation with the structural and hydraulic analysis needed to explain what the evidence shows.
A bridge failure routinely puts several of these in motion at once:
Fracture surfaces, bearings, gusset plates, and the collapsed geometry itself are the evidence. Clearing debris, cutting members for removal, or reopening a damaged span before documentation can destroy the proof of sequence and cause.
By comparing what the inspection and maintenance record shows was known against what the structural analysis shows the as-built structure could actually carry. A maintenance failure typically shows a documented defect — a rated section-loss finding, a known crack — that was not addressed within a reasonable interval. A design deficiency shows up as a member or connection that could not carry its rated load even in as-new condition. The two are not mutually exclusive, and separating them is usually the central question in the case.
Usually yes, if the fractured surface is preserved. Fatigue fracture surfaces carry beach marks and ratchet marks that point back to the origin site, and the origin often coincides with a specific weld toe, hole, or geometric discontinuity. What destroys this evidence is cutting through the fracture during debris removal or allowing the surfaces to corrode together, which is why securing the fractured member intact is the first priority.
Scour is the removal of streambed and foundation soil by flowing water, and it can undermine a pier or abutment without any visible sign above the waterline. Proving it after the fact relies on bathymetric surveys taken during and after the event, hydraulic modeling of the flood discharge, and comparison against baseline soundings from prior inspections — scour holes often partially infill afterward, so timing the survey matters.
It depends on where the failure mechanism traces to. A design deficiency implicates the engineer of record; a missed or understated inspection finding implicates the inspection contractor or the owning agency's program; a construction defect implicates the contractor. Bridge failures frequently involve more than one contributing factor, and allocating responsibility requires reconstructing the full chain from design through the most recent inspection cycle.
The load rating is the engineering statement of what the bridge can legally carry, and it is often the first document examined. An investigation checks whether the rating reflected the true as-built condition, whether known deterioration had been factored in, and whether the loads actually crossing the bridge, including any permitted overweight routes, exceeded what the rating allowed. A rating that was never updated after a documented condition change is a recurring finding in these cases.
Technical briefings from our work in this area.
Scour removes the material supporting a foundation during a flood and often refills the hole as the water recedes. The evidence is transient, which shapes how these investigations have to be run.
readA fatigue crack in a steel bridge member grows for years before it matters. The fracture surface records that history, and the inspection file records who had a chance to find it.
readAt 1:27 AM on March 26, 2024, the container ship Dali lost propulsion and struck the Francis Scott Key Bridge in Baltimore. The bridge — an open-deck truss design — collapsed in 13 seconds. Six construction workers died. The failure is now the most consequential structural-marine collision case in a generation.
readTell us what happened. We will triage it and connect you with the right expert — usually within one business day.