Distortion models for estimating human error probabilities

Human Reliability Analysis aims at identifying, quantifying and proposing solutions to human factors causing hazardous consequences. Quantifying the influence of gives rise error probabilities, whose estimation is a cumbersome problem. Since these are usually related other organisational or technological factors, it has been proposed apply probabilistic graphical models, such as Bayesian credal networks. However, can be problematic when conditional probabilities on missing data involved. While so far combine frequentist subjective approaches in general not robust small modifications dataset, this paper we propose an alternative based distortion which type imprecise probabilities. We perform comparative analysis, showing that our proposal consistent with previous studies while giving estimations.