Carroll High-hazard industries such as nuclear power and chemical process plants need to learn and improve without sole reliance on trial-and-error, because serious accidents are too costly and dangerous. These industries already place considerable attention and resources on learning from operating experience, including exchange of best practices, peer reviews, and self-assessment. Yet there are frequent failures to improve or to sustain improvement, either because problems are misunderstood, or because change is not implemented effectively. This research project focuses on developing a data base of several dozen incident review teams in four nuclear power plants and two chemical plants. These teams are composed of representatives from multiple groups such as operations, maintenance, and engineering, immediately following discovery of a serious problem or performance trend. Such team efforts are reserved for the most important and complex concerns each year. Each team is responsible for gathering and analyzing information and then making recommendations for changes in plant equipment, procedures, training, organization, and so forth. Using a series of questionnaires from team members and managers who are the sponsors or recipients of reports, and direct observation of teams in action, we address several interrelated issues at the group, individual, and organizational levels. At the group level, we would like to understand how to put the right expertise and cognitive abilities together in order to produce a more effective report and more learning among the team members that is diffused across the plant after the team has completed its work. At the individual level, we would like to understand how participation on an incident review team makes the team members more thoughtful about their own and others' work, and more complex in their thinking. At the organizational level, the issue is how plant improvement depends upon the nature of the team report and its reception by management and plant employees. We hope to develop new theories and practical information for management of high-hazard industries.
