Analysis of human error in occupational accidents in the power plant industries using combining innovative FTA and meta-heuristic algorithms

Occupational accidents are of the main issues in industries. It is necessary to identify the main root causes of accidents for their control. Several models have been proposed for determining the accidents root causes. FTA is one of the most widely used models which could graphically establish the root causes of accidents. The non-linear function is one of the main challenges in FTA compliance and in order to obtain the exact number, the meta-heuristic algorithms can be used. The present research was done in power plant industries in construction phase. In this study, a pattern for the analysis of human error in work-related accidents was provided by combination of neural network algorithms and FTA analytical model. Finally, using this pattern, the potential rate of all causes was determined. The results showed that training, age, and non-compliance with safety principals in the workplace were the most important factors influencing human error in the occupational accident. According to the obtained results, it can be concluded that human errors can be greatly reduced by training, right choice of workers with regard to the type of occupations, and provision of appropriate safety conditions in the work place.