A targeted risk prediction method based on statistical analysis of accidents in petroleum geophysical exploration

Abstract

Petroleum geophysical exploration (PGE) has a significant impact on the oil industry, but still trapped by fieldwork accidents. How to predict risks of accidents at worksites in a targeted way is still a challenge, but few researchers have focused on this issue because of the lack of data from the worksite. To address this problem, statistical analysis is carried out based on accident information acquired from a well-known oil corporation and a targeted risk prediction (TRP) method has been proposed in this paper. Firstly, an accident information statistic template (AIST) for key indexes extracted from accident investigation reports is established as the basis. Secondly, the combination of Chi-Square test and grey relational analysis (GRA) method is applied to determine the relationship between accidents’ causation variables and accident types. Chi-Square test is used to narrow the range of causation factors relating to specific accident types, whereas the GRA method is employed to help locate the accident type that most likely to occur under certain conditions. Finally, a TRP method of PGE fieldwork is proposed based on the statistical calculation results. With this method, fieldworkers can be aware of the overall risk at worksites as well as concentrating on preventing accident types that most likely to occur in a specific working condition before a fieldwork. An application study of this method has also been discussed. This method can be used to provide targeted pre-warning messages and corresponding prevention strategies before performing a PGE fieldwork. Thus can hopefully enhance the PGE fieldwork safety.