Assessing Human Performance Influencing Factors through LINMAP and Bayesian Belief Networks

Abstract

This study aims at identifying and ranking the performance influencing factors (PIFs) that cause error in human operations. The failure weight and rate for the tasks carried out by each operator were investigated. Assessing these factors reduces human error, hence increasing safety, efficiency and job satisfaction. The methods of linear programming technique for multidimensional analysis of preference (LINMAP) and Bayesian Belief Networks were used to investigate an aircraft tire manufacturing industry. All operators in the workshops were evaluated. Based on data analysis, the tasks of each operator were weighted and the potential error rate of each task was obtained. PIFs for each workshop were ranked and prioritized so that the factors having most influence can be easily distinguished and the tasks where the operators have the highest rate of failure, can be identified. The probability of human error was obtained and in a predictive model, it can be determined that when an error occurs, which factors are most influential in its occurrence. The method utilized does not include copious pairwise, exhausting and at times confusing comparisons for operators.