Great many of Accident Investigations are so superficial as to be useless

The intention behind the accident investigation is to find the underlying causes behind the accident, and then to amend relevant regulations and conditions to make sure that this accident do not happen again. Why is this well intentioned procedure so ineffective?

In SOLAS chapter I, Regulation 21, it is stated that each Administration undertakes to conduct an investigation of any casualty occurring to any of its ships "when it judges that such an investigation may assist in determining what changes in the present Regulations might be desirable".

The purpose of this regulation is to make sure that the same accidents do not happen again. But why leave it up to the various national Administrations discretion to decide when an accident investigation should be carried out?

In theory, the accident investigation should establish the direct and underlying cause and the IMO regime can then change the relevant regulations accordingly. But in practice, far too few reports are forwarded to IMO. According to "IMO News"; five years after IMO established the Marine Casualty and Incident Report Database, IMO requested Member States to provide information relating to 2.896 casualties, but only 1.676 was received. "IMO News" further commented that "the full value could not be obtained from many reports due to the lack of information provided. It was even evident that in some cases no investigation at all had been undertaken.

Furthermore, the probability of preventing reoccurrence of the accidents in the future is very remote, unless the lessons learned are passed on to the seafarers. At present, there seems to be no effective mechanism for getting such messages through to them.

Another problem stems from legal difficulties involved in the release of information, particularly where such information might reveal negligence or liability of the parties involved. Consequently, managers often demonstrate reluctance to divulge information necessary to reveal the causes behind the accident, which in fact is what the accident investigation is looking for.

The IMO instrument for accident investigation is the Assembly Resolution A.849(20), "Code for the Investigation of Marine Casualties and Incidents." The Resolution is aimed at providing a standard procedure to the investigation, in order to correctly identify the causes and even more important; to find the underlying causes of the accident.

Unfortunately, this Code has for many years been subject to revision by IMO's Flagstate Implementation subcommittee, heading a group of 13 flag States. Regrettably, this has created even more confusion, as several Administrations in the meantime have acted as if the Code was temporary "out of enforcement".

An example is of this is the Norwegian government's refusal of to allocate the necessary funding to The Norwegian Maritime Directorate, upon its request to carry out a full investigation of bulk carrier Rocknes' grounding in Norwegian waters in 2004, an event with many fatalities.

Looking at some of the publicly available accident investigation reports from the recent years, we can discern certain patterns. On the positive side, we may mention the loss of the containership MSC Napoli, which was abandoned after suffering a catastrophic failure of her hull in way of the engine room in January 2007. The accident investigation by the UK Marine Investigation Branch revealed a failure of the construction rules for buckling strength in way of the engine room, resulting in crucial amended design criteria.

The successful accident investigation seems to be when the cause of the accident is found and corrected by regulatory or rule amendment or design reinforcement. A common handicap is when a ship sinks, important evidence is lost with it, and usually such cases is closed without causal factors being revealed. An example of this is the bulk carrier MS Derbyshire, which was lost in the Pacific in 1980, without any trace. But the wreck site was found in 1998, and astonishing new findings resulted in crucial improvements of safety rules for bulk carriers.

In the extent possible, all shipwreck should be located and analysed to find the factual evident that caused the accident.

Another problem is that some accident investigation reports turns out to be very vague about the real causes and the sequence of events leading up to the accident, but reveal insted a long list of possible causes. It may be in the order of 40-60 different findings, which may or may not have contributed to the final accident.

This kind of accident investigation is more like a Risk Assessment exercise, which could be done at any time and much cheaper too. The Code in fact recommends that the accident investigation should analyse and comment on the causal factors, showing how they tie together, in order to reach logical conclusions,

An example of the typical "multiple cause accident" is the accident investigation after the loss of the Norwegian High Speed Craft Sleipner in 1999, where the commission came up with some 30 "probability recommendations" for regulation amendments. Some of them turned out to be a useful enhancement of general safety, such as improvement of life raft stowage and life jackets, but none of these matters was in fact the cause of the accident. The most probable cause was the malfunction or malpractice of the ECDIS navigational system, a matter which is still discussed within radar and nautical science publications.

As a conclusion, it is a sad fact that a large percentage of Accident Investigations are so superficial as to be useless. The reason is that when the real causes are not correctly identified, there is little or no return on the investment of the time spent doing them.

A more effective use of casualty investigations could help accidents at sea to be reduced. This is a problem in several industries. In order to make the learning from accident investigations more effective, several transportation bodies and research institutes in Sweden and Norway, including the Norwegian Maritime Accident Investigation Branch, have founded a research project, ACCILEARN (How to improve learning from Accident Investigations). Maybe the shipping industry have something to learn too.

Oppdatert: 29.06.2008 22:52
av Carina Grelland Harsem