Application of Hyperheuristics in Dynamic Environments

It has been shown that evolutionary algorithms enriched with heuristics are capable of finding promising solutions for dynamic optimization problems. The utilization of hyperheuristics for the maintenance of cooperation among such heuristics is an interesting topic. In this study, hypermutation, random initialization and using a diploid representation are chosen as the three basic techniques that enable evolutionary algorithms to cope with dynamic environments. Adaptive and self-adaptive hyperheuristics have been integrated in the system to control the progress of the lower level basic techniques. This paper aims to report the results of a series of experiments which were performed to test the performance of adaptive and self-adaptive hyperheuristics for the coordination of the specified basic techniques under different change conditions. The dynamic bit-matching problem is chosen as the test problem.