GA-based method for fuzzy optimal design of system reliability with incomplete FDS

Abstract

In this paper, we formulate a fuzzy nonlinear integer programming problem as an optimal design of system reliability with incomplete fault detecting and switching (FDS) that includes fuzzy numbers which allow the decision-maker to be more flexible, and solve it directly by keeping the nonlinear constraint by using an improved genetic algorithm (GA). The GA employs the variable penalties and the criterion of unfeasible chromosomes for improved evaluation function and improved arithmetic crossover. As a result, the improved GA increases the search efficiency in the solution space. We discuss the efficiency by comparing the proposed GA with traditional simple GA (SGA).