Hazard identification and prioritisation practices are very important for power plants to continue their operations without disruption. Systematic operational hazard analysis is not a very common practice at the Heavy Fuel Oil (HFO) based power plants in Bangladesh. Hence, a structured hazard evaluation framework can greatly benefit them to ensure their operational safety. This study has been conducted to identify and prioritise the operational hazards of the HFO-based power plants through using a hybrid multi-criteria decision-making (MCDM) approach in a fuzzy environment, and then, to explore the appropriate mitigation methods for the top-ranked hazards and to find the interrelationships that exist among the mitigation methods. First, the most common hazards in HFO-based power plants have been identified from the expert feedbacks. Then, a fuzzy analytical hierarchy process (FAHP) method has been used to determine the weights of the evaluation criteria and a fuzzy technique for order performance by similarity to ideal solution (FTOPSIS) method, has been used for the final ranking of the potential hazards. Afterwards, mitigation methods for the top 25 hazards have been identified and interrelationship among those mitigation methods has been explored through using interpretive structural modelling (ISM) and a matriced impacts croisés multiplication appliquée à un classement (MICMAC) analysis. The study finds that ‘explosion of high-pressure steam drum of the gas boiler’, ‘crankcase explosion and fire hazard due to oil pressure rise’ and ‘explosion of the compressed air reservoir’ are the top three hazards in the hazard ranking. ‘Standard operating procedure (SOP) and training’ have been found to be the most driving mitigation methods for the top-ranked hazards based on the ISM-MICMAC analysis. The findings of this study are expected to provide the managers of power plants with valuable insights, which can help them to prepare sustainable operational strategies to ensure the least hazardous work environment.
Read full abstract