Using Expert Models in Human Reliability Analysis-A Dependence Assessment Method Based on Fuzzy Logic

Abstract

In human reliability analysis (HRA), dependence analysis refers to assessing the influence of the failure of the operators to perform one task on the failure probabilities of subsequent tasks. A commonly used approach is the technique for human error rate prediction (THERP). The assessment of the dependence level in THERP is a highly subjective judgment based on general rules for the influence of five main factors. A frequently used alternative method extends the THERP model with decision trees. Such trees should increase the repeatability of the assessments but they simplify the relationships among the factors and the dependence level. Moreover, the basis for these simplifications and the resulting tree is difficult to trace. The aim of this work is a method for dependence assessment in HRA that captures the rules used by experts to assess dependence levels and incorporates this knowledge into an algorithm and software tool to be used by HRA analysts. A fuzzy expert system (FES) underlies the method. The method and the associated expert elicitation process are demonstrated with a working model. The expert rules are elicited systematically and converted into a traceable, explicit, and computable model. Anchor situations are provided as guidance for the HRA analyst's judgment of the input factors. The expert model and the FES-based dependence assessment method make the expert rules accessible to the analyst in a usable and repeatable way, with an explicit and traceable basis.

Comparing the treatment of uncertainty in Bayesian networks and fuzzy expert systems used for a human reliability analysis application

“…The idea behind the work in [28] and [29] is to build a computable model that represents the relationships among the relevant factors 5 (for the present application, relevant to the dependence between two events) and between these factors and the HEP. The potential benefits from the use of expert models in HRA (connected with possible increased transparency and repeatability of HRA analysis) have been discussed by the authors of the present paper in [27].…”

“…A discussion of the limitations of current approaches for human error dependence treatment is beyond the scope of the present paper (see [27] for more details). The idea behind the work in [28] and [29] is to build a computable model that represents the relationships among the relevant factors 5 (for the present application, relevant to the dependence between two events) and between these factors and the HEP.…”

“…The use of expert systems for HRA dependence assessment here is demonstrated on a working model proposed in [27] and [28] that builds on the qualitative rules suggested in THERP. The model does not include all factors that may impact the dependence level, but is deemed of enough complexity for the purposes of this demonstration.…”

“…The direct use of expert judgment to assess the qualitative dependence level has a number of weaknesses, including the traceability and repeatability of the results. In some cases, simplified approaches such as decision trees are used; however the rules represented by these trees do not have a transparent basis [26][27].…”

Human reliability assessment (HRA) in maintenance of production process: a case study

“…Deacon et al (2013) implemented a combination of expert judgment techniques for evaluating the risk during the evacuation opreration. Further, Podofillini et al (2010) proposed a fuzzy expert system (FES) formalism for dependence assessment in HRA, which captures the rules used by experts to assess dependence levels and incorporates this knowledge into an algorithm and software tool. Zio et al (2009) suggested HRA framework, which is based on a fuzzy expert system in which a set of transparent fuzzy logic rules is used to represent the relationship between the input factors and the conditional human error probability.…”