Assessing dependence in human reliability analysis using probabilistic linguistic term sets

Abstract

Dependence assessment is an important part in human reliability analysis (HRA). Normally, the judgments provided by the experts for dependence assessment are linguistic terms, and uncertainty could exist in the judgments. In this paper, a novel dependence assessment method based on the probabilistic linguistic term set (PLTS) is proposed to assess dependence among human actions. Firstly, the influential factors of the dependence assessment are identified, and their weights are determined using the analytical hierarchy process (AHP). Then, judgments provided by the analysts are presented using PLTSs. Next, the PLTSs from different analysts on a specific factor are combined, where their weights are determined using a new PLTS weight calculation method. Finally, the dependence level and the conditional human error probability (CHEP) are obtained by combining PLTSs of all influential factors. The proposed method can address the uncertainty in the dependence assessment in HRA, which inherently is a linguistic assessment process.