A synthetic risk assessment model based on AHP

Abstract

As an important part of system safety, risk assessment is an evaluation process for identified hazards. The corresponding risks should be evaluated to support decision-making. The risk, including the severity and probability of the potential accident, should be evaluated using an integrated index to facilitate risk ranking. In this study, the risk assessment is executed though the analytic hierarchy process (AHP) theory which allows synthesizing the severity and probability in evaluating a hazard. Using the AHP theory, the impact weights of the severity and probability can be determined by a pair-wise comparison matrix during which scales are used to describe their relative contributions to the risk. The severity and probability grades are defined to evaluate the severity and probability. The corresponding trapezoidal fuzzy numbers of these grades are also introduced. Finally, the risk of hazard is calculated using the impact weights of severity and probability. Comparing with the traditional risk assessment, the proposed risk assessment model considers the different impact weights of severity and probability to risk level and provides a more reasonable risk assessment result. An example of risk assessment in aviation industry is introduced and the comparison of the results between the proposed model and the traditional risk assessment is also made to demonstrate the engineering practicability and effectiveness of the new model in the risk assessment process.