Software multiple-fault localization using particle swarm optimization via genetic operation

Recently, spectrum-based fault localization approaches have been widely used for its fast and perform well for programs with only one fault.However, most of the existing methods do not consider the fact that the programs tend to have multiple faults. To address the above issue, we propose a Particle Swarm Optimization with genetic operation based Multiple-Fault Localization (PSOMFL). Our method models the software multiple-fault localization process as a search process for the particle swarm algorithm, which can quickly find the optimal solution in the multi-dimensional hyper-volume, and finally analyzes the optimal solution set to obtain the locations of multiple faults. We have implemented a prototype and conducted several experiments to compare PSOMFL against the existing fault localization approaches. The experimental results show that PSOMFL outperforms the compared methods and can reduce the costs by 5%-25% on average.