Test Case Prioritization Using Metaheuristic Search Techniques

Abstract

In this paper, Artificial Particle Swarm Optimization (PSO) inspired by real Swarm social–psychological tendency is used to solve time constraint prioritization problem-the techniques to prioritize the test cases that finds faults as early as possible, or maximize the rate of fault detection in the suite. The proposed technique is compared with three searches based metaheuristic approaches–(1) an ant-colony optimization approach, (2) Cuscuta search algorithm and (3) Hybrid Particle Swarm Optimization algorithm and two evolutionary metaheuristic- (1) Multi-Criteria Genetic algorithm technique which the fitness is APFD and (2) Multi-Criteria Genetic algorithm technique which the fitness is the proposed fitness multiplied by APFD and with five other non-search based prioritization techniques- (1) optimal, (2) random, (3) reverse, (4) untreated and (5) average faults found per minute algorithm based ordering. We investigate whether the proposed PSO metaheuristic outperforms existing prioritizing techniques in terms of APFD Score.