Stochastic rule-based decision support system for reliability redundancy allocation problem

Abstract

Reliability Redundancy Allocation (RRA) is one of the most important problems facing the managers to improve the systems performance. In the most RRA models, presented in the literature components’ reliability used to be assumed as an exact value in (0,1) interval, while various factors might affect components’ reliability and change it over time. Therefore, components reliability values should be considered as uncertain parameters. In this paper, by developing a discrete - continuous inference system, an optimization - oriented decision support system is proposed considering the components’ reliability as stochastic variables. Proposed DSS uses stochastic if - then rules to infer optimum or near optimum values for the decision variables as well as the objective function. Finally, In order to evaluate the efficiency of the proposed system, several examples are provided. Comparison of the inferred results with the optimal values shows the very good performance of the developed stochastic decision support system.

Keywords

• reliability redundancy allocation problem
• stochastic decision support system
• stochastic rule base