Multi-objective artificial bee colony algorithm in redundancy allocation problem.

This paper presents an empirical study of uncovering Pareto fronts by multi-objective artificial bee colony for redundancy allocation problem (RAP). Multi-objective artificial bee colony has been successfully applied in many optimization problems; however, a very little effort has been extended towards solving RAP. In this work, we have considered simultaneous optimization of the unavoidable objectives that are maximization of reliability, minimization of cost, and minimization of weight in a series parallel system, which leads to a multiple objective redundancy allocation problem (MORAP). The objective of this paper is to uncover true Pareto fronts populated with non-dominated solution sets as a solution to MORAP using multi-objective artificial bee colony algorithm (MOABC). Two MOABC algorithms have been developed and are inspired from the popular and established multi-objective genetic algorithms like Vector Evaluated Genetic Algorithm (VEGA) and Non-dominated Sorting Genetic Algorithm II (NSGA II). We named these two algorithms as MOABC-I and MOABC-II, respectively. From the experimental results, we visualize that the approximation of true Pareto front by MOABC-II is better than Pareto front obtained through MOABC-I. Further this resultant Pareto fronts are supervised by two inherent multi-criterion decision making (MCDM) methods like Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and Analytical hierarchy process (AHP) to reach at a definite goal.