Comparing SLIM, SPAR-H and Bayesian Network Methodologies

Abstract

Human factors always affect maintenance performance, and in some cases, it’s critical to systems availability and reliability. Despite such importance, in so many cases, there’s no human reliability method applied to analyze maintenance tasks in order to understand better human factors influence in maintenance performance. There are several human analysis methodologies and regarding human factors, SLIM (Successes Likelihood Methods), SPAR-H (Standardized Plant Analysis Risk-Human Reliability Analysis Method) and Bayesian Net take into account such factors and may be a good approach to minimize human error. In order to propose a human reliability methodology to analyze maintenance tasks taking into account human factors, a case study about turbine star up tasks will be carried out. Therefore, different human reliability methods will be performed based on specialist opinion. Finally, the human error probability as well as drawbacks and advantages from different methods will be discussed to get a final conclusion.