Task Based Evaluation in Error Analysis and Accident Prevention

Abstract

After a system failure that is attributed to human error, an accurate accounting of the factors that led to the error is critical for redesign efforts and any resulting litigation. Without data on the design components or environmental factors that contributed to the error, the human is indicted, either legally or in the press, no system redesign is conducted and the failure will be repeated at the cost of additional damage and/or lives. However, rarely is the human at fault because of intentional negligence. There are always design or environmental factors that interacted with the person's training, knowledge, motivation and other factors that can be affected by the system. Accident investigations must be conducted to identify these factors and guide redesign efforts. An even better situation would be if this same testing occurred before the system was released to the market. This requires prediction and simulation of the people who will use the system, the tasks that will be attempted, the motivations of system users, and the environmental factors that will interact with the system. Testing systems under controlled laboratory conditions is generally insufficient to identify potential sources of human error. This paper describes the critical components of task-based evaluation and presents a case study to illustrate the consequences of failure to use TBE.