Artificial Bee Colony Optimization for Multiobjective Quadratic Assignment Problem

ABSTRACT: Excellent ability of swarm intelligence can be used to solve multi-objective combinatorial optimization problems. Bee colony algorithms are new swarm intelligence techniques inspired from the smart behaviors of real honeybees in their foraging behavior. Artificial bee colony optimization algorithm has recently been applied for difficult real-valued and combinational optimization problems. Multiobjective quadratic assignment problem (mQAP) is a well-known and hard combinational optimization problem which is used in modeling of several assignment and scheduling problems. Benchmark mQAP instances are already solved near optimally using competitive metaheuristics and dedicated local search algorithms, but there is no absolute winner of these competitions in the sense that while a particular algorithm is quite successful for a kind of mQAP instance, it exhibits poor performance on the others. In this study, we test the performance of artificial bee colony optimization algorithm over multiobjective quadratic assignment problem. Experiments have shown that the new heuristic was effective and efficient to solve hard mQAP instances. Keywords: Multi-objective optimization, Artificial Bee Colony, Bees Algorithm, Multiobjective Quadratic Assignment Problem. …………………………………………………………………………………………………………………………