The 2024 collapse of Baltimore’s Francis Scott Key Bridge sparked an intense debate in engineering circles about what could be done from a civil engineering perspective to make bridges more impact-resistant. Some insisted that the weight of a container ship such as Dali made the outcome of the Key Bridge inevitable and that the ship’s loss of power was largely to blame for the tragedy. However, it’s worth noting that the Key Bridge was fracture critical, so any event that takes out one pier could cause cascading failure.
As SmartBrief learned from Jason Hastings, vice chair of the American Association of State Highway and Transportation Officials’ Committee on Bridges and Structures, AASHTO’s current vessel-collision design provisions were developed when ships were much smaller and the available data more limited. He noted that at the time of the collapse, the committee was already reviewing how those criteria should evolve.
The conversation continued last week during a session at the American Concrete Institute’s Fall convention in Baltimore, where engineers, researchers and code specialists examined new approaches to modeling, materials and risk evaluation.
Force, energy and risk in design
Michael Knott of Moffatt & Nichol traced how the collapse of the Sunshine Skyway Bridge collapse in Tampa, Fla., in 1980 led to AASHTO’s first Guide Specification for Vessel Collision Design of Highway Bridges in 1991. Those guidelines eventually became part of the Load and Resistance Factor Design Specifications. Knott noted that while hundreds of vessel-bridge contacts occur each year worldwide, a small number can cause major damage, often with severe consequences.
Ship size and energy capacity have increased sharply since the 1990s. Container vessels that once carried 8,000 twenty-foot equivalent units now exceed 20,000. Because kinetic energy increases with the square of velocity, even small changes in vessel speed can greatly magnify the impact force transmitted to a pier. Current LRFD provisions classify bridges by consequence level: Critical and essential crossings are designed for a one-in-10,000 annual probability of collapse. Typical bridges target one-in-1,000 probability. Meeting those benchmarks requires modeling both the peak force and the total energy absorbed through deformation, cracking and damping within the concrete and fender systems, Knott explained.
Finite element analysis has become a key tool for estimating those effects, but Knott cautioned that the accuracy of results depends on reliable data for vessel mass, stiffness and structural response. In some cases, Knott added, operational measures such as tug assistance or reduced navigation speed can provide equivalent risk reduction at far lower cost than new construction.
Reliability and system behavior
Andrzej Nowak of Auburn University shared a reliability-based calibration method for evaluating vessel-collision limit states. His analysis treats the bridge, vessel and navigation channel as an interacting system and uses statistical techniques to describe uncertainty in impact conditions and structural capacity. Target reliability indices ranged from about 3.5 for typical bridges to more than 6 for critical crossings, similar to values used for other low-probability, high-consequence events such as earthquakes.
Nowak’s study suggests that human error and mechanical failure remain the dominant causes of collisions, while factors such as flood current, visibility and channel geometry influence exposure. By adjusting AASHTO’s load and resistance factors to align with site-specific data, he suggested that engineers could achieve more consistent levels of safety without excessive conservatism.
Evaluating existing concrete bridges
Fadi Oudah of Dalhousie University described a probabilistic framework for assessing harbor bridges built before vessel-impact provisions were introduced. The approach combines random-field finite element modeling with Monte Carlo simulation to represent variation in material properties and deterioration, including reinforcement corrosion and section loss. The resulting resiliency index quantifies the probability of different damage states and helps bridge owners prioritize maintenance or retrofitting.
The framework allows for simplified screening and detailed nonlinear modeling, depending on data quality and structural importance. Oudah’s findings suggest localized deterioration in reinforced concrete piers can significantly reduce impact capacity, even when the overall geometry remains unchanged.
Current design practice
Emily Ullmer of WSP reviewed the development of AASHTO’s collision provisions and the analytical methods commonly used today. She described three general design approaches: deterministic, based on representative vessel loads; probabilistic, which incorporates collision frequency and failure probability; and cost-effectiveness, used when full retrofits are not practical.
For new bridges, engineers iterate between span layout and risk modeling until the target probability of collapse is met. For existing structures, the analysis determines an equivalent return period that indicates whether further mitigation is needed. Ullmer emphasized the importance of Automatic Identification System data for building accurate models of vessel size, speed and traffic frequency at specific waterways.
Application, policy and research
Adrienne Crider of Modjeski and Masters discussed how some analytical tools are being implemented in the field. Many bridges constructed before the 1980s were not designed for vessel impact, she explained, so owners must identify where improvements will have the greatest effect. Crider described a phased process that begins with vulnerability screening, continues with refined dynamic analysis for high-risk sites and leads to targeted physical countermeasures such as dolphins, concrete cells or pier strengthening.
She also highlighted ongoing studies of real-time monitoring systems that use sensors to detect vessel proximity or pier movement. Such data could support both emergency response and long-term calibration of design models. Coordination among bridge owners, the Coast Guard and insurers remains a challenge. In many cases, she said, bridge owners carry most of the financial risk even when accidents result from navigation error. Some international organizations are considering higher design impact loads or shorter return periods, though cost implications are still being evaluated.
_____________________________________________________________________________________
This is part 3 of SmartBrief’s recap of the American Concrete Institute’s 2025 fall convention. Click here for part 1 and here for part 2. For more insight into the latest news and trends affecting the concrete industry, subscribe today to ACI SmartBrief and SmartBrief for Civil Engineers.