Selection of the optimal method for analysis of accidents in petroleum industry using fuzzy ANP and TOPSIS multi - criteria decision methods

Occurrence of accidents in the oil industry induces to the irreparable damage to human life and high value property. Understanding the causes of accidents takes place with analysis. One of the most important steps in accident analysis is a conscious selection of accident analysis method due to the existence of high variety of methods and various effective criteria. This study was carried out with the aim of selecting the optimal method for analysis of accidents in petroleum industry.
A multi–criteria decision model was developed which was comprised of the Analytic Network Process to obtain the criteria weights (which was used for selection of the accident analysis method) and fuzzy TOPSIS for ranking of the accident analysis methods.
Based on the results, 8 criteria and 21 sub criteria were used for selection of the accident analysis method. The factors of multiple levels cause identification capability, training need and levels of analysis with final weights of 0.375, 0.168 and 0.088 (respectively) had the most important weights in method selection. Based on fuzzy TOPSIS ranking of accident analysis method, tripod beta was selected as the optimal method and other methods such as BOW tie and MTO were in subsequent priorities.
The results showed that there were several criteria (with different weight) for selecting the optimal method of accident analysis. Based on important of criteria incorporated in the present study, the optimal method for petroleum accident analysis is Tripod beta. Incorporation of Tripod beta in combination with the other methods may be resulted in numerous benefits in accident analysis.