Dynamic risk analysis for oil and gas exploration operations based on HFACS model and Bayesian network

Abstract

Oil and gas exploration (OGE) is an important part of the petroleum industry. How to predict and prevent risks of OGE accidents in a targeted way is still a challenge, while few researchers have focused on this issue because of the lack of data from the worksite. Therefore, the accident cases for decades of years in a globally operated OGE company are collected as a basis, a dynamic risk analysis method combining the human factors analysis and classification system (HFACS) and the Bayesian network (BN) has been proposed in this article. First, identify the accident cases, analysis, and divided the risk factors into four levels and 27 classification categories, and setup a HFACS-OGE model. Then, map the factors to the BN and identify the correlation of the secondary level factors. Finally, setup a BN model with the Genie software. Through posterior probability analysis and sensitivity analysis, we conclude the major and sensitive factors leading to OGE accidents. According to the conclusion, administrators could decide on corresponding management and control methods. This method can hopefully provide an effective quantitative basis for the identification and prediction of the factors that may lead to OGE accidents.