Welcome to the PSAM 18 Abstract Status page.
Abstract ZX112Full Paper + Presentation
Updating the Dependency Equations Proposed in THERP Using Experimental Data
Dependency analysis is a method used in human reliability analysis (HRA) to adjust the failure probability of a subsequent action by considering the impact of a preceding action. The Technique for Human Error Rate Prediction (THERP) HRA method originally proposed an approach to dependency analysis with five distinct levels of dependence: Zero Dependence, Low Dependence, Medium Dependence, High Dependence, and Complete Dependence. Each dependency level applies a specific equation to adjust the failure probability of the subsequent action. For example, the equation for Low Dependence assumes 5% of the distance ratio between the basic human error probability (i.e., Zero Dependence) and 1.0 (i.e., Complete Dependence) and uses it to provide conditional failure probabilities. However, the distance ratios for the dependency equations were assumed without any supporting background. Furthermore, the THERP handbook recommended that these ratio values be updated in the future. Nevertheless, the THERP equations, with their current assumptions regarding the distance ratios, have continued to be used as is.
This study aims to update the THERP dependency equations using experimental data. The data used in this study was collected through the Simplified Human Error Experimental Program (SHEEP) framework, an HRA data collection framework proposed by Idaho National Laboratory (INL). The SHEEP framework focuses on simplified simulators and non-expert operators to complement full-scope HRA data collection studies. The SHEEP database includes human error and performance data collected from 36 student operators and 36 actual operators using the Rancor Microworld simulator, developed by INL, and the Compact Nuclear Simulator, developed by the Korea Atomic Energy Research Institute (KAERI). Through data analysis, this study ultimately aims to answer the following questions.
• How can dependency distance ratios be estimated and THERP dependency equations be updated using experimental data?
• How do dependency methods or PSFs differentiate distance ratios?
• Can we reduce the number of dependency levels?
• Do the dependency factors currently used in HRA truly affect the dependencies between human actions?
• Isn’t there new dependency factors observed from experimental data?
• Can we use other data sources (e.g., License Event Reports) to update the dependency equations?
← Check another abstractThis study aims to update the THERP dependency equations using experimental data. The data used in this study was collected through the Simplified Human Error Experimental Program (SHEEP) framework, an HRA data collection framework proposed by Idaho National Laboratory (INL). The SHEEP framework focuses on simplified simulators and non-expert operators to complement full-scope HRA data collection studies. The SHEEP database includes human error and performance data collected from 36 student operators and 36 actual operators using the Rancor Microworld simulator, developed by INL, and the Compact Nuclear Simulator, developed by the Korea Atomic Energy Research Institute (KAERI). Through data analysis, this study ultimately aims to answer the following questions.
• How can dependency distance ratios be estimated and THERP dependency equations be updated using experimental data?
• How do dependency methods or PSFs differentiate distance ratios?
• Can we reduce the number of dependency levels?
• Do the dependency factors currently used in HRA truly affect the dependencies between human actions?
• Isn’t there new dependency factors observed from experimental data?
• Can we use other data sources (e.g., License Event Reports) to update the dependency equations?