MULTI-CRITERIA OPTIMIZATION: MAXIMIZATION OF ASYSTEM RELIABILITY ESTIMATE AND MINIMIZATION OFTHE ESTIMATE VARIANCE

摘要

Problem formulations and descriptions of optimization approaches are presentedrnwhen there are multiple design objectives; (1) to maximize an estimate of systemrnreliability, and (2) to minimize the variance of the reliability estimate. This is arnsound and risk-averse formulation that reflects the actual needs of system designersrnand users. Designs with the highest estimate of system reliability may actually be arnpoor choice for a risk-averse applications because of an associated high variance.rnMore realistic approaches to the problem are presented to incorporate the riskrnassociated with the reliability estimate. System design optimization then becomes arnmulti-criteria optimization problem. An example is solved using nonlinear integerrnprogramming methods. The results indicate that very different designs are obtainedrnwhen the formulation is changed to incorporate minimization of reliabilityrnestimation variance as a design objective.