Incidents are a source of information about management and control, and a good basis for further improvements. Operating companies can and should learn from incidents. However, most information that can be obtained from incidents is hardly used as a result of time constraints. Proper incident analyses according to proven and detailed methodologies are only conducted for incidents with severe impacts, or the potential of severe impacts. Luckely incident rates are low in the E&P sector, but that also means that detailed incident analyses are rare. The E&P sector as a whole could therefore benefit from sharing incident data, enlarging the learning database. Although there is some sharing, and benchmarking against lagging KPI's like LTI frequencies, there is no sharing of consistently classified data about root causes, direct causes and failing control measures. It is this information operating companies would benefit most from, since it directly indicates where action is required.

In this paper the results are presented and discussed of a review, analyses and classification of 478 fatal accidents in the period 2005 – 2009, as reported bij members of the International Association of Oil & Gas Producers (OGP). Information about incidents of OGP members is publicly available on the OGP website. The information from the incidents have resulted in statistics of direct causes, root causes and lacking or failing control measures, including some trends. This paper has the intent to initiate a step change in the thinking about, and in taking action with respect to analyzing and sharing available incident data in industry on a common analysis and classification approach, in order to enlarge the learning database and enable real risk-based management.

Since this review has been based on publicly available information, it is considered a mirror of E&P safety performance from an independent, third-party perspective. All information was already available. The strength of this study is the consistent analyses on the basis of a single analysis model, independent of the source of information. The outcomes presentated provide a good basis to focus management and control, and even to focus regulatory inspections by government bodies.

You can access this article if you purchase or spend a download.