Identification of the human factors contributing to maintenance failures in a petroleum operation.

ABSTRACT

OBJECTIVE: This research aimed to identify the most frequently occurring human factors contributing to maintenance-related failures within a petroleum industry organization. Commonality between failures will assist in understanding reliability in maintenance processes, thereby preventing accidents in high-hazard domains. BACKGROUND: Methods exist for understanding the human factors contributing to accidents. Their application in a maintenance context mainly has been advanced in aviation and nuclear power. Maintenance in the petroleum industry provides a different context for investigating the role that human factors play in influencing outcomes. It is therefore worth investigating the contributing human factors to improve our understanding of both human factors in reliability and the factors specific to this domain. METHOD: Detailed analyses were conducted of maintenance-related failures (N = 38) in a petroleum company using structured interviews with maintenance technicians. The interview structure was based on the Human Factor Investigation Tool (HFIT), which in turn was based on Rasmussen's model of human malfunction. RESULTS: A mean of 9.5 factors per incident was identified across the cases investigated.The three most frequent human factors contributing to the maintenance failures were found to be assumption (79% of cases), design and maintenance (71%), and communication (66%). CONCLUSION: HFIT proved to be a useful instrument for identifying the pattern of human factors that recurred most frequently in maintenance-related failures. APPLICATION The high frequency of failures attributed to assumptions and communication demonstrated the importance of problem-solving abilities and organizational communication in a domain where maintenance personnel have a high degree of autonomy and a wide geographical distribution.