An Expert System for Solving Multi-objective Decision Making Problems: MPSO

Multi Particle Swarm Optimization (MPSO) is a modified form of Particle Swarm Optimization (PSO) which includes various sub-swarms instead of single swarm, thereby balancing the exploitation and exploration. The optimal solution termed as global best (gbest) is achieved by passing the best fitness value obtained from child swarm and further progression of gbest is achieved by parent swarm. Travelling Salesman Problem (TSP) is an NP hard problem aim is to find the minimum distance for a given cities by traversing exactly one’s to reach the final destination. In this paper MPSO techniques is used for solving multiobjective constraints problems using the proposed Boundary value Analysis (BVA) techniques. Analysis of the proposed techniques is done by comparing with the standard dataset taken from the TSP lib. The analysis is performed in terms of computation time, iteration, optimal value based on best case, worst case and average case, further more analysis is performed based convergence diversity and average convergence diversity. Form the analysis is it very clear that proposed techniques out perform well within a minimum computation time and minimum iteration, in case of particle 30. In future we plan to implement the proposed techniques in cloud computing environment for job scheduling problems.