Solving the Quadratic Assignment Problem with the modified hybrid PSO algorithm

Abstract

In this paper a particle swarm optimization algorithm is presented to solve the Quadratic Assignment Problem, which is a NP-Complete problem and is one of the most interesting and challenging combinatorial optimization problems in existence. A heuristic rule, the Smallest Position Value (SPV) rule, is developed to enable the Continuous particle swarm optimization algorithm to be applied to the sequencing problems. So, we use SPV to the QAP problem, which is a discrete problem. A simple but very efficient Hill Climbing method is embedded in the particle swarm optimization algorithm. We test our hybrid algorithm on some of the benchmark instances of QAPLIB, a well-known library of QAP instances. This algorithm is compared with some strategies to solve the problem. The computational results show that the modified hybrid Particle Swarm algorithm is able to find the optimal and best-known solutions on instances of widely used benchmarks from the QAPLIB. In most of instances, the proposed method outperforms other approaches. Experimental results illustrate the effectiveness of proposed approach on the quadratic assignment problem.