Some Empirical Insights on Diagnostic Performance of the Operating Crew in a Computer-Based Advanced Control Room

Abstract

As computer-based design features are adopted in main control rooms of nuclear power plants, a human reliability analysis (HRA) method dealing with the effects of these design features on human behavior is needed. This article provides experimental results of human diagnostic performance characteristics in a computer-based, full-scope, dynamic simulator to inform some insights on developing an HRA method for a computer-based advanced control room. In comparison to the performance time for diagnostic actions, it showed more or less faster performance in the computer-based control room with a computer-based emergency operating procedure (EOP) for an event scenario with an apparent diagnostic symptom than in the conventional control room with a paper-based EOP, but it is also revealed that the diagnosis time is highly dependent on the situational characteristics of simulated events. Regarding the aspect of human error occurrence, a decision maker showed the potential for leading to a wrong conclusion regarding the plant state when he makes a situational assessment or makes a decision based on abnormal information by himself without communicating or consulting with other operators. Finally, regarding the aspects of error recovery, it showed that the error recovery potential becomes much higher for the advanced control room than for the conventional control room due to the information sharing and access capability of the advanced control room between and for all the crew members. It is expected that an HRA method for an advanced control room environment should adequately reflect these characteristics of human behavior in a computer-based control room. (C) 2010 Wiley Periodicals, Inc.