Uncertainty of human error and fuzzy approach to human reliability analysis

Human error is regarded as a critical factor in catastrophic accidents such as disasters at nuclear power plants, air plane crashes, or derailed trains. Several taxonomies for human errors and methodologies for human reliability analysis (HRA) have been proposed in the literature. Generally, human errors have been modeled on the basis of probabilistic concepts with or without the consideration of cognitive aspects of human behaviors. Modeling of human errors through probabilistic approaches has shown a limitation on quantification of qualitative aspects of human errors and complexity of attributes from circumstances involved. The purpose of this paper is to investigate the methodologies for human reliability analysis and introduce a fuzzy logic approach to the evaluation of human interacting system's reliability. Fuzzy approach could be used to estimate human error effects under ambiguous interacting environments and assist in the design of error free work environments.