DUTCH non-profit safety outfit Maritime Research Institute Netherlands is nearing the end of a three-year research programme on the causes of container losses and will shortly deliver its findings to the International Maritime Organization.

Rob Grin, senior project manager at Marin — as the institute is known for short — was speaking at the International Union of Marine Insurance conference in Berlin.

“Containerships are not intrinsically safe but are designed to be safely operable,” he told delegates.

The work is being carried on under the rubric TopTier, and is supported by 41 shipping entities, including many big name box carriers and classification societies.

Just 0.0006% of containers carried are lost, with a rolling average over an unspecified period of 1,482 boxes, as far as can be attained from publicly available information.

Indeed, 2023 proved a very good year in this respect, with only 221 boxes lost, Grin said.

But consequences can be substantial when things do go wrong, and TopTier aims to bring container losses down to zero.

Parametric rolling has been identified as a major cause of container losses, and TopTier has developed a roll risk estimator based on Microsoft Excel spreadsheets, and produced instructional videos to make it easy for seafarers to use.

Another issue is that the actual strength of containers and lashings can be less than the maximum safe working load.

The actual mass of containers can be higher than the declared mass. While it is widely believed that this situation obtains only in 1% of cases, TopTier has identified a terminal where 13% of containers were overweight in a given week.

Sometimes containers are not stowed according to the stowage plan. Sometimes the motions of ships exceeds the most severe design case, with actual forces exceeding those calculated by lashing software.

Mike Roderick, a partner at law firm Clyde & Co, also explained why unseaworthiness rarely succeeds as a defence for marine insurers.

While warranties of seaworthiness and cargoworthiness are contained in the Marine Insurance Act 1906, they are frequently overwritten in practice. There is no warranty of seaworthiness under a time policy, for instance.

If underwriters wish to run a defence under s.39(5) of MIA, they must show not just unseaworthiness, but privity and ensuing loss.

In practice the problem is proving privity, which is to say that the shipowners knew about the unseaworthiness.