A novel approach to repair time prediction and availability assessment of the equipment in power generation systems using fuzzy logic and Monte Carlo simulation

Abstract

This research propounds a general theoretical and practical solution for the simulation of the repair time of equipment and forecasting equipment preventive maintenance, availability, repair rate and availability indicators. The experience of experts was used to estimate the time to failure and repair of equipment. In this way, the critical years of the equipment were determined in terms of the duration of equipment downtime by evaluating and checking the annual availability. The process of repair time was simulated by designing a neuro-fuzzy system to predict repair time. Also, Monte Carlo simulation was used to calculate realistic annual availability, time-dependent repair rate and other availability factors. A case study on an oil pump was carried out for the validation of the proposed method. According to the results, applying preventive maintenance in the optimal time intervals of 190–200 days had a significant effect on the equipment uptime, availability, and the number of its additional periodical inspections. Also, the minimum availability of the equipment was predicted to be 96% and the maximum to be 99%. Finally, the case study results helped determine the general graphs of time-dependent repair rate and availability for engineering equipment.