Charlie looks into the recent incident at Francis Scott Key Bridge in Baltimore and discusses the potential impacts of the accident.
Often when I write my bulletin, it is about the latest cyber incident and what we can learn from it. However, last week, the crash of the MV Dali into the Francis Scott Key Bridge and its subsequent collapse, resulting in the loss of six lives, caught my interest and prompted me to write about it. This is a classic business continuity and emergency response incident, and I thought I would examine it through the lens of a couple of risk management concepts: black swan events and Reason’s Swiss cheese theory.
Black swan events, popularised by Nassim Nicholas Taleb, are rare events that have a major impact when they occur. Traditional risk management focuses on identifying and mitigating risks that can be predicted based on historical data and past experiences. Black swan events, by definition, fall outside this realm. They are inherently unpredictable and unseen. They usually fall outside the realm of normal expectations and historical data. When they occur, they have significant impacts and lasting consequences. What is also interesting about them is they contain a high element of hindsight bias. When the event occurs, people quickly jump to the conclusion that the event was inevitable even if nobody was predicting it beforehand.
The containership MV Dali hitting one of the piers of the Francis Scott Key Bridge has all the characteristics of a black swan event. The event caused the collapse of an iconic bridge, which will have a major impact on traffic in the area as it is regularly used by 35,000 commuters a day. The bridge will take months if not years to rebuild. The Baltimore port is key for imports of cars to the USA, and according to The Times newspaper, the port being unable to function and send and receive cargo will cost the local economy approximately $15m a day. Six members of a construction crew died as they were working on the bridge when it collapsed. The ship will have to be refloated, once it can be untangled from the collapsed bridge. I suspect it, and its cargo, will be held in port for weeks if not months while the incident is investigated. The cargo was destined for Sri Lanka, and so there may be a knock-on impact on their businesses and economy. According to Lloyd’s, this event might be the biggest ever marine loss in the history of the insurance industry as the event was a ‘fully insured risk event’, meaning that the Francis Scott Key Bridge, the ship, the vessel’s cargo, and the port, were all covered. There could also be secondary claims from the ships stuck in Baltimore harbour waiting to leave. Estimates for the calamity are between $1 billion and $3 billion. The cost of the Costa Concordia cruise ship hitting a rock in Italy was approximately $1.5 billion.
According to The Times, ‘An analysis of 503 bridge collapses in the US between 1989 and 2000 found that only 2 percent were caused by ship collisions’. So, bridge collapses are very rarely caused by ships. All the commentaries I have read on the incident have have said that there was no warning of a possible incident or any calls for the bridge to be better protected so therefore, in terms of the hindsight bias element of black swan events, this was true.
This incident also conformed to Reason’s Swiss cheese theory, where there was a requirement for a number of different events to align for the event to occur. In this instance, the first element was that the ship was having electrical issues before the incident, and for the two days it was in Baltimore port, repairs were being conducted on the ship’s electrical system as it kept tripping out. Location and timing of the incident were all important in causing it to crash into the bridge and collapse. The ship lost power and had to wait for the backup generators to kick in. The very minute the main power failed, the crew lost control of the ship and were not able to regain control and steer away from the bridge supports. If this incident happened in port or at sea, then there may have been little or no impact. It happened at the worst possible location.
Most bridges which go over shipping lanes are highly protected with uprights to prevent ships hitting them. Either they have small islands surrounding them or they have ‘dolphins’ in place which steer the impact of a ship hitting away from the bridge. The Francis Scott Key Bridge uprights did have some protection, but they would only have withstood the impact of a small boat. The bridge was constructed 50 years ago, and the designers may not have predicted the huge increase in the size of ships and the subsequent force if they hit the uprights. The only mitigation was that the incident occurred in the middle of the night rather than at peak rush hour, so the casualty count was much less than it could have been. All these elements had to align for the incident to occur and have the impact it did.
The location and timing were very important in causing this incident and it reminded me of the factors which caused the Selby rail crash in the UK in 2001, in which 10 people died. A Land Rover came off the M62 motorway, slid down a bank, landed on the tracks below, blocking the path of an oncoming train which caused the crash. The reason the crash had so many casualties was that as one train crashed into the Land Rover, there was a passenger train going in the other direction, and the two trains crashing into each other was the cause of the casualties. One of the elements which contributed to the crash was that the driver, Gary Hart, was up all night on dating apps and fell asleep, but the other contributing factor was the place he went down. The bank was the only short stretch of motorway without a barrier which would have kept the car on the motorway. The deaths were caused as a result of the two trains passing each other while the driver was trying to inform the police. If they were 5 minutes earlier or later, it was unlikely for the incident to occur. If the crash took place at any other location the only casualty was likely to be the driver, so as seen in the Baltimore incident, a number of instances need to be aligned to cause the incident.
When you study academic theories, you sometimes wonder about their general applicability to real-life incidents and if they only work for a few that the authors of the theories have based their theory on. In this case, I think it is very clear that this is a black swan event and that Reason’s Swiss cheese theory applies.
