No abstract

A traditional view of incidents is that they are caused by shortcomings in human competence, attention, or attitude. It may be under the label of ''loss of situational awareness, '' procedure ''violation,'' or ''poor'' management. A different view is that human error is not the cause of failure, but a symptom of failuretrouble deeper inside the system. In this perspective, human error is not the conclusion, but rather the starting point of investigations. During an investigation, three types of information are gathered: physical, documentary, and human (recall/experience). Through the causal analysis process, apparent cause or apparent causes are identified as the most probable cause or causes of an incident or condition that management has the control to fix and for which effective recommendations for corrective actions can be generated. A causal analysis identifies relevant human performance factors. In the following presentation, the anatomy of a radiological incident is discussed, and one case study is presented. The contributing factors that caused a radiological incident are analyzed. Underlying conditions, decisions, actions, and inactions that contribute to the incident are identified. This includes weaknesses that may warrant improvements that tolerate error. Measures that reduce consequences or likelihood of recurrence are discussed.

4 _ T D $ D I F F ] At Los Alamos National Laboratory (LANL), there are several nuclear facilities, accelerator facilities, radiological facilities, explosives sites, moderate-and high-hazard non-nuclear facilities, biosciences laboratory, etc. The Plutonium Science and Manufacturing Directorate (ADPSM) provides special nuclear material research, process development, technology demonstration, and manufacturing capabilities. ADPSM manages the LANL Plutonium Facility. Within the Radiological Control Area at TA-55 (PF-4), chemical and metallurgical operations with plutonium and other hazardous materials are performed. LANL Health and Safety Programs investigate injury and illness data. In this study, statistically significant trends have been identified and compared for LANL, ADPSM, and PF-4 injury/illness cases. A previously described output metric is used to measures LANL management progress towards meeting its operational safety objectives and goals. Timelines are used to determine trends in Injury/Illness types. Pareto Charts are used to prioritize causal factors. Data generated from analysis of Injury/Illness data have helped identify and reduce the number of corresponding causal factors.

No abstract

No abstract