Improved Z-number based fuzzy fault tree approach to analyze health and safety risks in surface mines

Surface mining is vulnerable and subject to a wide range of risks, requiring extensive risk analysis to ensure mine health and safety (MHS). Fault tree analysis (FTA) is a graphic representation tool for conducting safety and reliability analyses by modeling causal chains that lead to failures. However, conventional FTA cannot deal with uncertain and imprecise information. Therefore, in order to handle the uncertainty arising from lack of complete information and to enhance the reliability of qualitative judgment of experts, the concepts of Z-numbers and fuzzy theory are combined with FTA. The proposed approach used expert elicitation to comprehensively analyze MHS risks related to machine/equipment, environment, and workplace. Through causal inquiries of the FTA, 8 undesired events and 65 underlying basic events are explored and analyzed, taking into account the probability of occurrence of all basic events. Further, a sensitivity analysis is performed using Fussell-Vesely Importance and Risk Reduction Worth Methods to verify the model and examine how each of the basic events contributes to the occurrence of any undesirable incident. Results reveal that issues associated with blasting, dust, and explosive fumes are the most probable incidents to occur among the undesired events. The main basic events causing MHS risks result from non-implementation of regulations, staff incompetence, improper safety perimeter setting, and explosive calculations. This study assists practitioners in making risk management decisions and implementing corrective measures. The proposed approach can also be applied to investigate similar risk factors in different industries. •Fault Tree Analysis is conducted for surface mine health and safety related risks.•An improved z-number based fuzzy Fault Tree Analysis system is developed.•Eight undesired events and 65 underlying basic events are explored and analyzed.•Sensitivity analysis is performed to verify the model.•Blasting, dust, and explosives fumes are the most probably undesirable events.