Engineers and scientists react to the Francis Scott Key Bridge collapsing in Baltimore. 

1) Dr Stergios Mitoulis, head of Structures Research Group, University of Birmingham

“The Francis Scott Key Bridge, known originally as the Outer Harbor Crossing until it was renamed in 1977, or simply the Key Bridge or Beltway Bridge, was a steel arch-shaped continuous through truss bridge spanning. The bridge is 1.6 miles (2.6km) long and the central part that failed was a 366m long continuous truss.

"A heave ship collided on the bridge and the pylon was not able to take the excessive collision/dynamic load and the bridge collapsed due to progressive collapse. Removing one support redistributes loads in the structure, akin to the way loads shift when standing on one leg. Thus even though it might look as a disproportionate consequence, the pylon collapse would lead to collapse of the bridge in all cases – this is not something that we can design our bridge decks for (ie, complete failure of a pylon).

“We don’t know the condition of the foundations, whether footings were scoured or not, whether reinforcement inside RC pylons/columns was extensively deteriorated due to corrosion, and the type of the foundation and the condition of the pylons/foundation/deck. We don’t know the material properties and strength and the condition (corrosion, connections and inspection records/strengthening efforts in the past).

"It is common to design bridges and their columns and pylons such that an impact/collision load is envisaged. However, we don’t know if collision loads were taken into account during the design in the 1970s and if the impact force exceeded the design load. Also, we don’t know the speed of the ship, its mass, and route, angle of attack and location of the impact, and the potential displacements that this impulse/impact caused.

“All these are investigations that will take place in the aftermath of bridge collapse. These forensic investigations might take one to two years as lab testing and collapse scenarios will have to be analysed to come up with plausible/reasonable simulations of what might have happened today. Barge/ship/tanker represents 2-3% of all bridge failures across the world. More than 60% of failures are due to flooding and other climate change related stressors.

“The condition of the pier and its foundation is the most important aspect of the forthcoming investigations. Bridge metallic truss looks in good condition and painted (from Google images) but further investigations are required to determine their structural integrity during the incident.

“There was no way to protect the bridge, even if there was a warning system in place. If a ship like this collides on with any bridge it may take it down.”

2) Dr Roberto Gentile, assistant professor in catastrophe risk modelling, University College London

“In analysing this event, investigators will likely establish the level of vessel-impact load the bridge was designed for (according to the design code in place at the time). They would compare this load with the one resulting from this event, which reasonably is low probability event. But the considerations can be much more general.

"One could compare the load from this event with the design load in today’s code regulations. Ultimately, the question is 'what is the margin of safety of bridges designed today (according to the minimum standards) against such rare events?'

3) Ian Firth, independent consultant, fellow and past president of the Institution of Structural Engineers

“It is almost impossible to design a bridge pier to withstand this kind of impact. Therefore, we tend to design impact protection measures to prevent it from happening instead.

“Dolphins or other vessel impact protection devices in the water are commonplace since the Sunshine Skyway collapse in 1980. But this bridge was built in the 1970s, so the design would not have incorporated these devices at that time. The fact that a vessel can veer off course and hit the pier is the reason to design vessel impact protection systems so that a large vessel cannot hit the critical bridge support.

“The footage shows there are small dolphins, (the small round objects visible in the film), each side of the bridge piers – these have not prevented the vessel hitting the pier.

“A new bridge design would probably be a cable-stayed bridge with a much larger span, moving the supports well away from the navigation channel and into shallower water.”

4) Prof Jin Wang, professor of marine technology, Liverpool John Moores University

“While the exact causes leading to this unfortunate tragedy remain to be seen, loss of propulsion of the ship seems to be a likely one. When the propulsion power is lost in a ship, the manoeuvrability is lost because propellers, thrusters, etc. would not be operational. In such cases the ship may drift randomly, particularly in harsh weather conditions. In this Baltimore bridge collapse, when the accident happened, the sea seemed to be calm with the wind at a speed of less than 2km/h. As a result, the ship should not drift away randomly and violently if loss of power was the cause.

"To avoid the shipping drift randomly to lead to possible collisions with other objects, the best way of mitigating possible consequences is to anchor the ship and also to ask emergency response and rescue services (eg, tugs or similar ships) to assist the ship. The accident happened just before 1.30am on March 26, 2024. Such an early time may not have helped the ship operators find the best way of mitigating possible consequences.

“The Dali container ship involved in this accident is less than 10 years old with gross tonnage of 95,000 tonnes. It is considered to be of an average-sized container ship with 22 crew members on board. The ship has operated in the US-Asian route. The ship was in a healthy state. It would be interesting to know how the autopilot was used during the early hours of March 26, 2024.

"The AIS (Automatic Identification System) that is a coastal tracking system currently used on ships, should have useful information recorded to help find out the exact causes of this accident. The speed of the vessel when the collision happened, should also be accurately found from the AIS system or the VTS station nearby. This would also help find the reason why the whole bridge collapsed due to the collision.

“There are clear general guidelines for vessel navigation under the bridge. There are modern technologies such as marine radar to help deck officers manoeuvre the ship to go under bridges. As a result, it is rare to see accidents of this nature. While accidents of this nature are rare, there have been reported cases in Europe and the Far East recently. For example, not long time ago, in the early morning (local time) of February 22, 2024, a cargo ship rammed into a bridge on the Pearl River near Nansha in China, plunging five vehicles including a public bus into the river.

“The exact causes leading to the accident can only be found through a formal accident investigation. While it is possible to have equipment failures, human error cannot be ruled out. It is likely that human error played a role in this accident. Human error can be reduced through more resilient design with greater redundancy so that if one system stops working, another takes over.

"Crew training may also be a contributor to this tragedy. Crew members should be trained to be able to respond to challenging circumstances during ship operation. It is likely that the operators on board the vessel did not respond in the best way they could have done. However, this remains to be seen till the full independent investigation is carried out.

“An independent accident investigation should take place in terms of what happened, how it happened, why it happened and what can be done to prevent it happening again. This may potentially lead to change of operational rules of ships from the International Maritime Organization (IMO), the United Nations specialised agency with responsibility for the safety and security of shipping and the prevention of marine and atmospheric pollution by ships.

“The shipping industry traditionally operated in a reactive regime, meaning that design and operational rules and regulations were modified following the occurrence of serious accidents. With the introduction of Formal Safety Assessment (FSA) by the IMO, the marine industry has increasingly moving towards a proactive goal-setting risk-based regime. This is particularly useful with new technologies and new risks continuously emerging.

"FSA is a framework of hazard identification, risk estimation, identification of risk control measures, cost benefit analysis and making decisions. It can be applied to any maritime system design and operation with a view to avoiding or minimising risks in a proactive manner. Any possible new regulatory or rule changes to be introduced by the IMO will be justified through conducting a FSA.”

5) Chris Goldsworthy, chief executive of Institute of Marine Engineering, Science and Technology

“What is required now is a full and detailed investigation that avoids any inherent blame culture and instead focuses on the real, and as always, multiple root cause identification, including full systems and relevant system design changes prompted, encompassing the human element across all aspects, with the aim to prevent reoccurrences and to enhance safety for all, at sea and on shore.”

6) Alan Hayward FREng CEng FICE FIStructE, retired bridge engineer

“Collapse of the 804m-long three-span section of the bridge appears to have been caused by the massive container cargo vessel DALI impacting and demolishing one of the two main bridge piers. This took away vital support from the steel truss superstructure with its suspended roadway, such that the whole three-span section then suffered from catastrophic structural failure and collapsed into the river below..

“The vessel appears to have impacted virtually head on with the pier, implying that the force applied would have been very significant. The force would depend on a number of factors including weight of the vessel, its speed, and distance over which its travel was arrested. Assuming a vessel dead weight of perhaps 100,000 tonnes the potential force could well exceed 1,000 tonnes, and is likely to be well beyond the capacity of even robust bridge piers. I have no details of the pier strength or its foundations.

“I would have expected to see a system of ‘dolphins’ and/or fendering, adequately secured into the river bed. Such a system would be aligned on both sides of each main pier with the purpose of deflecting errant vessels from collision with the piers. Collision protection would generally be designed to resist glancing or sideways impact forces, much less than from head-on impacts. Other measures would comprise warning systems including navigation lights etc.

“From the photos I can see what appears to be a solitary dolphin-type structure on each side of the two main piers. If so, these look to be inadequate to deflect anything other than small vessels. I do not know what measures were planned or installed when the bridge was opened in 1977, but the sizes and weights of cargo vessels have increased enormously to the present with the globalisation of container sea transport.

“As far as reconstruction of the bridge is concerned it would seem preferable to adopt a much longer main span so that the risk of ship collision is much reduced. The bridge type would be of cable-stayed or suspension form. Piers would be protected and designed to resist practicable specified forces from sideways deflected vessels.”

7) Dr Marina Bock, lecturer in structural engineering, Aston University

“The Francis Scott Key Bridge is a metal truss bridge with a suspended deck, and from videos shown by the media it seems that the vessel has hit a main pier of the bridge. The main piers rest on soil underwater and they are part of the foundations of the bridge.

“This type of bridge is not designed to redistribute loads in the event of a main pier collapsing and therefore videos show there is a progressive collapse of the bridge (bridge elements fail after other bridge elements had failed).

“Maybe if the vessel had hit a small section of the suspended deck the bridge would have been able to survive the collision but not a main pier.

“I am assuming one of the first line of investigation will be to understand why the vessel could not avoid the pier when there were no other vessels around.”

8) Dr Sotirios Argyroudis, associate professor in infrastructure engineering, Brunel University London

What we know so far

“The bridge is 2.6km long, built in 1977. The supporting structure – the bridge piers – is made of reinforced concrete. The metallic truss of the superstructure was continuous.

“The collapse occurred when a ship collided with one of the bridge piers, leading to the failure of the metallic truss. This exemplifies a common mode of failure known as progressive collapse. In such cases, the removal of one boundary of the bridge support – here, the pier – causes a redistribution of loads that the metallic truss cannot effectively bear, resulting in its progressive collapse. The primary cause of this failure was the significant forces exerted on the pier due to the collision with the ship.”

What we don’t know

“We don’t yet know the recent condition of the bridge, including the condition of the metallic truss and the condition of the pier’s foundation. Any of these may have been deteriorated since 1977 – for example, due to scour effects.

“We also don’t yet know the conditions during the collision, including the speed and mass of the ship, or angle of impact, which are needed for calculating the force exerted by the moving ship on the bridge pier.”

Any additional information you can provide based on what you know

“Approximately 42% of bridges in the USA are more than 50 years old, potentially posing safety risks in certain instances.

“Ageing infrastructure is particularly vulnerable to deterioration caused by factors like corrosion and environmental conditions. Moreover, exposure to various hazards such as earthquakes, floods, hurricanes and human-induced stressors like collisions and explosions exacerbates their risk of failure. Compounding the issue, current traffic loads often exceed those for which these bridges were originally designed. Conducting resilience assessments of our infrastructure is crucial for informing future investments in infrastructure improvement.”

To illustrate this point 3, Dr Argyroudis has provided the following table from other academics’ research:

Distribution of causes of 503 reported bridge collapses during the period between 1989 and 2000 in the United States (Wardhana and Hadipriono 2003). Collapses resulting from ship collisions account for 2% of the total reported incidents.

9) David Knight, bridge expert and specialist adviser to Institution of Civil Engineers

“Safety and robustness is crucial when you are designing and building a bridge. This incident is a tragic example of what can happen when things go wrong.

“In the aftermath, the most important thing we can do is investigate any navigational errors that contributed to the accident and what if anything went wrong in terms of the bridge’s structure and general state of repair.

“One thing to note is that ship impact is a key consideration in modern, large span bridges in busy navigation areas.

“The bridge industry is very good at sharing knowledge, so whatever outcome we can learn from this incident will contribute to designing and building even safer bridges.”

10) Dr Raffaele De Risi, senior lecturer in civil engineering at University of Bristol’s School of Civil and Mechanical Engineering

“The Francis Scott Key Bridge in Baltimore was a long transportation infrastructure spanning over the Baltimore Harbor with 11 reinforced concrete piers based in water, nine auxiliary spans over water and a main steel arch-shaped continuous through truss bridge over three spans, supported by four piers, two frame-like piers at the two end of the truss and two inverted ‘V’ shape piers at the centre. The four piers were simply supporting a continuous, beautiful structural system.

“However, as often happens, such excellent systems lack structural redundancy with respect to the boundary conditions. This is natural for such a massive structure spanning over a river. Therefore, if one of the supports fails, the entire system follows.

“In the case of this bridge, unfortunately, one of the inverted ‘V’ shaped piers was hit by a 229m-long container vessel. A vessel impact is categorised in structural engineering as ‘Accidental load’, and sea waterway vessels with lengths between 50m and 300m may have a mass between 3,000 tonnes and more than 100,000 tonnes. Such massive vessels can have a catastrophic effect on piers if there is a lack of protection against impact. In addition, the pier was designed in the early 1970s for accidental loads that may have been lower with respect to modern ship traffic.

“Of course, our thoughts are with the people affected, their families, and the whole local community. In fact, the bridge, hosting more than 11 million vehicles annually, was undoubtedly a vital traffic link; its damage will make local travel and communications more difficult and the local community, unfortunately, less resilient.”

11) Dr Mehdi Kashani, associate professor in structural and earthquake engineering, University of Southampton

“I saw a video of the bridge collapse due to ship impact. Please not that my judgment is purely based on what I saw in the videos. Normally the bridge piers in rivers are either designed for ship impact or a ship impact barrier is designed and built around the pier to avoid ship impact. It is very similar to providing a safety barrier on the road to avoid car impact on bridge piers.

“I do not know the history of this bridge, but it looks like an old bridge that was designed neither for ship impact nor had any ship impact barrier to avoid the problem. I am not sure if there was any recent structural assessment on this bridge. Normally, in the case of old bridges if bridge engineers perform structural assessment they will investigate a scenario of ship impact and what damage it might cause. At this point the robustness of the bridge is key to avoid progressive collapse as we saw in the videos.”

12) Emeritus Professor Toby Mottram, structural engineer at the University of Warwick

“Historically, there have been instances where sections of bridge structures collapsed following collisions with marine vessels. Between 1960 and 2015 there have been 35 major bridge collapses worldwide due to ship or barge collisions.

“Regarding today’s Baltimore Key Bridge incident, the speed of the ‘Dali’ ship at the time of impact remains unknown. With an overall length of 300 metres and a width of 48.2 metres, the significant momentum of this massive cargo vessel, especially when laden with cargo, would have been considerable upon impact with the reinforced concrete pier.

“It’s evident that the pier couldn’t withstand the impact energy, leading to its failure and subsequent collapse of the steel truss and reinforced concrete deck superstructure. The extent of the damage to the bridge superstructure appears disproportionate to the cause, a matter for future investigation.

“When the steel truss road bridge was designed and built in the 1970s (opened in 1997), there likely wasn’t a requirement to consider disproportionate collapse and structural integrity. It’s conceivable that the piers weren’t designed to withstand the magnitude of today’s ship impacts, as vessels like the ‘Dali’ weren’t navigating the port of Baltimore during that era.

“Despite meeting regulatory design and safety standards of the 1970s, the Baltimore Key Bridge may not have been equipped to handle the scale of ship movements seen today. However, modern navigation technologies should have prevented the ship from striking the pier. Investigating what failed on the ship to cause the fatal impact will be a priority.

“The social and economic repercussions of this disaster are anticipated to be significant and prolonged.”

13) Dr Masoud Hayatdavoodi, expert in civil engineering and naval architecture at the University of Dundee’s School of Science and Engineering

“Coastal bridges are typically designed to withstand loads from vehicles, trucks, trains, and other sources above the deck, depending on their intended use. Factors such as seismic activity, wind loads, and the loads on bridge foundations are also taken into account during the design process. In recent years and following major storms, including those in the US, some coastal bridges have collapsed due to the impact of large waves and currents from below. This is now also a factor for consideration in bridge design.

“In areas where shipping is permitted, calculations are made to determine the maximum allowable elevation of ship from the water level. Unlike on the ground, this height varies in water, depending on factors such as water density and the cargo carried by the ship. In this case, it seems that the ship has collided with one of the columns of the bridge. Bridges are not designed to withstand lateral loads from ships on their columns. While they may be able to withstand lateral loads comparable to water currents and wind loads from a small boat, the force exerted by the impact of a container ship is significantly larger.

“There is no question that the bridge would collapse due to the impact on the columns. Where shipping is allowed under a bridge, there is typically a clear path for ships to pass safely, with a clearance margin. In this instance, there must have been a reason for the ship to deviate from its intended path and collide with the column.”

14) Dr Lee Cunningham, reader in structural engineering at University of Manchester

“The bridge is relatively modern having been constructed in the 1970s and it would be expected that some provision for ship impact around the support piers would have been accounted for in the design. The piers or support columns are primary load bearing elements of the bridge and any structural failure of these, particularly at the main central span, would lead to collapse.

“A vessel’s mass and velocity are key factors in the level of impact force generated and there is an economic and practical limit to what level of impact force can be designed for. Similarly, the direction of impact is also an important factor and design assumptions for this would likely be based on the position of the dedicated navigation channel.”

15) Dr Andrew Barr, research fellow in Department of Civil and Structural Engineering, University of Sheffield

“The collapsed part of the Key bridge is a continuous truss bridge, which means that it was constructed from a single long steel truss over the three main spans, rather than having hinged connections at the intermediate supports (piers). This style of construction is very efficient in terms of the amount of steel required, as loads can be shared between different parts of the structure. In the central span of the bridge the truss is arched, with the road deck suspended underneath on cables.

“The collision of a vessel as large as the Dali container ship will have far exceeded the design loads for the slender concrete piers that support the truss structure, and once the pier is damaged you can see from the videos that the entire truss structure collapses very rapidly.

“This is an example of what engineers call progressive collapse, where the failure of one structural element leads to the failure of neighbouring elements, which can’t support the new loads placed on them. In this case, the collapse of the pier caused the now unsupported truss above it to buckle and fall. Because this is one continuous truss, the loads are redistributed – the truss pivots around the surviving pier support like a see-saw, temporarily lifting the northern span into the air before the high tension forces cause this to fail too, and the whole truss collapses into the water.

“The video of the collapse will be incredibly useful to the teams involved in assessing the causes of the failure and how it progressed, as this would otherwise have to be done by careful analysis of the destroyed structure and comparison with modelling. The video doesn’t show any obvious structural deficiencies with the bridge, but it will not have been designed to survive a head-on collision with such a large vessel.

"Bridges in shipping lanes are sometimes designed with strong, stout piers, or additional protective structures around the piers to prevent ships from coming into contact with the bridge structure. It doesn’t appear that the Key bridge had either of these features, although it is also very likely that the size and design of the vessels passing under the bridge has changed considerably since it was completed in 1977.”

16) Prof Barbara Rossi, associate professor of engineering science, University of Oxford

“The bridge has received a huge impact force on one of its supporting structure. The supporting structure appears to be made of reinforced concrete. Two A-shaped concrete pillars above water connected with a horizontal beam form one part of the substructure, above water.

“I do not know anything about how the bridge is founded below the water level, but most probably on a set of piles (one can see the pile cap).

“The impacting force must have been immense to lead to these massive (concrete) structures to collapse, leaving the superstructure without one of its supports.

“As far as one can see in the video online, there appears to have been a very large transversal displacement  (perpendicular to the bridge, caused by the impact) which have caused the arch-shaped riveted steel truss to lose one of its supports. Following the collapse of the arch, the entire structure loses its stability (approaching spans).

“We should not speculate around if such huge impact forces should have been taken into account at design stage.”