A fuzzy causal relational mapping and rough set-based model for context-specific human error rate estimation

Safety is of paramount importance in high-risk systems. Safe and reliable operation of a system depends mainly on its key component, 'the human'. Human error is pivotal in an accident sequence and the rate of error is instrumental in the analysis of accidents which focuses on the development of interventions. However, assessing the rate of human error is a big challenge. Human performance and the rate of error are governed by the context of work. This study discusses a fuzzy rule-based causal relational mapping approach to investigate the variability of the human error rate with context. Different contexts are mapped onto the human error rate estimated from the reported accident cases. This mapping develops a causal relational diagram which can be used for predicting the human error rate in any context. Such information is useful to identify problems of areas and to develop safety countermeasures.