Feature: Boxes of known unknowns

One could be forgiven for thinking that what needs to be known about the millions of containers shipped around the world each year is indeed known. After all, since the terrorist attacks of 9/11 12 years ago, containers and their contents are supposed to have been subject to a battery of security checks that include intelligence-based screening, X-ray scanning and physical inspection.
More than 50 of the biggest ports around the world are now part of the Container Security Initiative (CSI) under which American borders have been effectively extended to foreign territory in a bid to prevent terrorists using US-bound containers to smuggle radioactive bombs or bomb-making materials.
The US Customs and Border Protection (CBP) also screens 100% of the cargo manifests now required to be submitted 24 hours in advance. It claims it has the capacity in personnel and equipment, including radiation detectors, to scan 99% of all containers arriving in the US by sea but recently admitted only 4% were actually scanned for radiation.
Plans to require 100% scanning of US-bound containers in overseas ports now appear to have been put on hold until at least next year as a result of doubts over the effectiveness of the technology, concerns over the likely disruption to trade and potential damage to relations with major trading partners including the European Union.
It was perhaps an acknowledgement of the importance to world trade of keeping the flow of containers – around 130 million TEUs were moved last year – as smooth as possible.
Keeping tabs on those millions of containers and their contents is a problem familiar to the industry itself which believes that not only is the true nature of the contents of many containers not known but that their true weight is also unknown.
These gaps in knowledge, the industry believes, compromise safety but filling the gaps is not proving easy. The industry is hoping the Cargo Information Notification System (CNIS) database of container incidents, set up in 2011, will prove an effective early warning system and trend-spotter.
Analysis of a year’s worth of incidents has confirmed some industry beliefs but also challenged some assumptions. It is estimated, for example, that over 20% involved cargo misdeclarations and many of these involved “dangerous” cargoes. At the same time, however, a third involved cargoes loaded in Europe and North America, areas previously thought to have been less likely to give cause for concern.
Misdeclarations can also involve understating the weight of the container and its contents, although the degree to which excess weight is responsible for incidents involving ships and road and terminal vehicles is hotly debated.
Last month through-transport insurer, the TT Club, which carried out the CNIS incident analysis, said was there was little doubt that the true weight of a “high percentage” of those 130 million containers could not be accurately known. Attempts, however, to introduce independent verification of weight will have to overcome resistance from sceptical shippers and embrace all the stages of a container’s movement, including road and rail.
The results of the CNSIS analysis also showed, the Container Owners’ Association (the database’s neutral host) said, the fears of the liner industry that the nature of the cargo it carried was “largely unknown” were reasonable.
Any relief containership owners and their insurers might have felt at the revelation that only 8% of reported incidents involved fires or explosions will have been tempered by their awareness that serious incidents continue to occur.
Last year’s MSC Flaminia incident, in which three of the crew died and the rest were forced to abandon their blazing ship, had revived fears that a mis-declared volatile cargo may have been responsible. German casualty investigators were, however, forced last month to suspend publication of their report that might have dispelled or confirmed such fears.
The investigators had been unable to begin their work in earnest until two months after the incident, delayed in part by the “time-consuming process” of finding a port of refuge for the disabled ship which was eventually accepted by Germany.
Pinpointing the origin of fires among stacks of containers, however, is never easy. In one of the few available casualty reports Danish investigators could only say one that hit the Charlotte Maersk in July 2010, damaging 160 containers, was “probably” caused by cans of methyl ethyl ketone peroxide. The potentially explosive chemical has been linked with several other fires, including one on the containership Cape Fresco a week after that on the Charlotte Maersk.
The report, published last year, also revealed the Danish containership was carrying 190 containers containing goods classed in varying degrees as dangerous.
At the same time as the MSC Flaminia was being readied for dry-docking in a Romanian shipyard last month, the fate of another, larger and younger containership was raising even more serious questions about safety.
The dramatic splitting in two of the MOL Comfort and the subsequent sinking of its aft section immediately prompted a demand for quick answers that cannot be readily met.
Speculation that overweight containers may have been a contributory factor has been based on previous incidents, although in the most quoted case – the MSC Napoli in 2007 – the casualty report by the UK’s Marine Accident Investigation Branch (MAIB) cited a number of other factors, including the structural strength of the vessel and its speed in heavy weather.
The MAIB report, however, said discrepancy in container weights was “widespread” and due in part to shippers deliberately under-declaring to reduce costs. It added pre-embarkation weighing was “essential” if stresses on containerships were to be “accurately controlled”.
As a former US Secretary of Defense once said about weapons of mass destruction, “There are known unknowns and there are unknown unknowns”, a phrase that might equally apply to containers, their contents and their true weights.
In the hunt for the needle in the container stack, reducing the number of known unknowns to a more manageable level may be the more realistic goal. The unknown unknowns may have to wait a bit longer.
Source: BIMCO