Modeling Risk with the Flight Operations Risk Assessment System (FORAS)

摘要

The Icarus Committee of the Flight Safety Foundation (FSF) has been examining ways to improve flight safety since the committee's inception in the early 1990s. The initial focus has been on improving flight safety with a decidedly human factors orientation. However, in the FORAS effort, the early work attempted to quantify safety as a metric that a flight safety department could use to monitor and thus improve the overall level of safety. When we applied Reason's (Reason, 1990) "paradigm" to our thinking, we shifted to the measurement of risk, as reported by Schwartz (1998). The purpose of this paper is to provide an update on the status of FORAS. During the 52nd International Aviation Safety Seminar (IASS) in Rio de Janeiro, McCarthy (op. cit.) provided an annual update to the Flight Operations Risk Assessment System (FORAS). A detailed technical presentation of FORAS was presented by Hadjimichael et al. (1999). The system is intended to be a measurement system that can determine the relative risk of an accident or incident. By relative risk, we mean that the system output is not an absolute measure of accident risk, but rather a number that is guaranteed to increase as the risk of a situation increases. Thus, while all flights have a very low risk of accident, flights can be compared to identify those flights where accident risk is greater. In our thinking, the analysis would be conducted before a potential or latent event occurs, by examining quantitatively a set of risk contributors, such as weather, crew, and airport conditions. By so doing, an airline or flight squadron could make changes in safety practices that put the organization in a situation that would make accidents less likely to occur. FORAS is not a means of preventing accidents; rather it is a means of quantifying situations that increase accident/incident risk. To this end, FORAS provides an opportunity to focus on combinations of factors that increase accident risk, as opposed to a system for predicting accidents.