Risk analysis of supply vessel operations

Abstract

The Norwegian Petroleum Safety Authority has recently raised a concern that allision incidents between visiting vessels and their relevant offshore installations may lead to major accidents.There have been 26 such allisions on the Norwegian Continental Shelf during the last 10 years.On the other hand, existing models for calculating the risk associated with supply vessel allisions are mostly simple and do not include all the factors that influence risk explicitly.

The aim of this study was therefore to investigate this problem area further and to develop an improved risk model.

Most of the incidents over the last 10 years were considered as minor, where the vessel strike the installation while working alongside, and where the consequences were negligible. However,several head-on incidents were also occurred while the supply vessel was approaching the installation. The consequences were considered as critical since the vessel strike the installation at relatively high speed. In this thesis, allision scenarios relevant for offshore supply vessel have been identified based on findings from allision investigation reports and literature studies. Detailed study has then been done on the head-on allision scenarios and a model for calculation of the allision frequency has been developed.

The model is based on the Barrier and Operational Risk Analysis (BORA) approach, which eases the incorporation of human and organizational factors into the model. A barrier block diagram was built that depicts the major barrier systems proposed to prevent the initiating event from developing into an allision. The initiating event is ‘supply vessel being on straight course towards installation’. The event sequences were shown using an event tree. Fault trees were developed to further analyze each barrier system. Thereafter, the most significant risk influencing factors were identified for each basic event in the fault trees.

Finding relevant input data to the model seems to be challenging. However, possible sources of input data to the model are proposed in the thesis. The functionality of model is validated basedon recent allision cases.

Due to time and resource limitation quantification of the model was not performed. Therefore, further research on this is recommended. In addition, a model for allision of supply vessels while working close to offshore installations is not developed in this thesis and needs to be taken as further work.