Complexity: Application to human performance modeling and HRA for dynamic environments

Abstract

Although the argument has been made quite strongly and perhaps convincingly that advanced, highly automated control rooms consisting of advanced digital information and control systems will be simpler to operate and maintain; the case can be made that this potential benefit will only be true to the extent that we successfully control the amount of complexity present for the operating crew. Borrowing from complexity theory and human reliability analysis literature, we review the aspects of complexity that should be considered, discuss the implications, and review the extent to which we find current human reliability analysis (HRA) methods sensitivity to complexity in advanced digital environments. Next, we identify 3 complexity factors related to human performance and propose a rating scale approach that allows the analyst to more clearly allocate complexity levels for HRA purposes. These factor scores are used to form the basis of a composite complexity score (CCS) that can be used to support HRA. Although one method, the Simplified Plant Analysis Risk Model Human Reliability Analysis (SPAR-H), is selected for purposes of mapping the 3 complexity elements, the results are thought to support the qualitative and quantitative portions of most HRA methods.