Application of heuristic methods for conformance test selection

Abstract

In this paper we focus on the test selection problem. It is modeled after a real-life problem that arises in telecommunication when one has to check the reliability of an application. We apply different metaheuristics, namely Reactive Tabu Search (RTS), Genetic Algorithms (GA) and Simulated Annealing (SA) to solve the problem. We propose some modifications to the conventional schemes including an adaptive neighbourhood sampling in RTS, an adaptive variable mutation rate in GA and an adaptive variable neighbourhood structure in SA. The performance of the algorithms is evaluated in different models for existing protocols. Computational results show that GA and SA can provide high-quality solutions in acceptable time compared to the results of a commercial software, which makes them applicable in practical test selection.