Tank explosions on large oil tankers and FPSOs
What is the probability of an explosion in the oil storage tanks of an FPSO? As far as we know, there never has been such an event, but the fatal explosion in the pump room on the FPSO Cidade de São Mateus  implied that it could happen. Offshore operators need to assess the risks of tank explosions in order to design against them. One way of estimating the probability is to use experience from a larger group of similar vessels, such as the world-wide fleet of trading oil tankers. There have been many explosions on such ships, in part because there are many more of them at sea. Their oil tanks are roughly similar to those on FPSOs.
So what is the frequency of an explosion in the cargo tanks of an oil tanker? We recently published an analysis of such events .
We chose data on large oil tankers, defined as those over 80,000 dwt, as these are similar in size to FPSOs. There are roughly 1200 such ships in service around the world, and over the period 1980-2013 we found 88 cases of fire and/or explosion in or near the cargo tanks. Of these, 48 were serious casualties that required repairs before the ship could continue trading.
The figure below shows the trend in the tank explosion frequency, i.e. the number of tank explosions each year divided by the number of ship in service. Clearly there have been fewer accidents recently, and the number is now so low that it is difficult to estimate the frequency, since there have been only two events in the last five years.
The reduction since the mid-1980s is probably due to the progressive introduction of inert gas systems, segregated ballast tanks and double-hulls. By 2010 these were fitted to the entire fleet, but there have been so few accidents since then that it is not yet possible to obtain robust results from this period alone. Using the population of double-hull tankers alone, we estimated a frequency of 3.9 x 10-4 per tanker year, i.e. an annual probability of about 1 in 2600 for each ship. After comparing alternative approaches, we estimated an uncertainty range of roughly a factor of 2 higher or lower.
Is this a realistic prediction for the future? The trend plot shows steady improvement over the last decade, which suggests the frequency will become lower still. However, recent regulations have resulted in a very high proportion of the fleet being new tankers, whose accident frequency is likely to increase as the vessels age. So it is possible that the trend will reverse. Positive trends have reversed before, notably in the late-1990s. This is why we prefer a relatively cautious approach, and do not attempt any extrapolation of the trend.
Would the frequency be any different on FPSOs? We reviewed the differences and concluded that some differences would tend to increase the frequency while others would reduce it. Until a more comprehensive analysis is available, we chose to avoid arbitrary adjustments and use the trading tanker data unmodified to estimate the frequency on FPSOs. A frequency of 3.9 x 10‑4 per FPSO year implies a tank explosion on average every 2600 FPSO-years. Cumulative experience with FPSOs to date is less than this, so the estimate is consistent with the absence of tank explosions to date. The new frequency is slightly lower than previous estimates , due to the use of more recent data.
In conclusion, we have shown that, on trading tankers, cargo tank explosions are rare and have been getting more infrequent. Despite this, we would not be surprised if the frequency increased somewhat as the fleet of double-hull tankers grows older. On FPSOs, storage tank explosions are as yet unknown, but trading tanker experience seems to provide a reasonable method of estimating their probability. Analysis of their causes and consequences could help evaluate the benefits of risk reducing measures.
 Brazilian Navy Directorate of Ports and Coasts, “FPSO CIDADE DE SÃO MATEUS Explosion Followed by Flooding with Casualties, February 11th 2015” Maritime Accident Safety Investigation Report.
 OGP (2010), “Storage Incident Frequencies”, Risk Assessment Data Directory Report 434-3, International Association of Oil & Gas Producers, March 2010.