Application of a multi-objective particle article swarm optimization technique to solve reliability optimization problem

This study investigates the applicability of multi-objective evolutionary algorithm (MOEA), namely the Particle swarm optimization incorporating crowding distance (MOPSO-CD) in solving the reliability optimization problem of a complex system where the two conflicting objectives maximization of system reliability and minimization of system cost are considered. The simulation results shows that MOEA used in this paper (MOPSO-CD) is able to generate a well distributed Pareto optimal set in a single run for a decision maker (DM) to choose from. DM could select the most convenient optimal solution according to his/her level of satisfaction in a posterior decision environment. The efficiency of MOPSO-CD in solving this problem is demonstrated by comparing its result with those of simulated annealing (SA), non-equilibrium simulated annealing (NESA) and cellular evolutionary strategies (CES).