Parameter Estimation of Software Reliability Growth Models by Particle Swarm Optimization
Alaa F. Sheta, Prince Abdullah, Bin Ghazi · 2007
Abstract Building software reliability growth models (SRGM)for predicting software reliability represents a chal-lenge for software testing engineers. Being able topredict the number of faults (failure) in the softwareduring development and testing processes helpssignificantly in specifying/computing the softwarerelease day and in managing project resources (i.epeople and money). In this paper, we explore the useof Particle Swarm Optimization (PSO) algorithm toestimate SRGM parameters. The proposed methodshows significant advantages in handling variety ofmodeling problems such as the exponential model(EXPM), power model (POWM) and Delayed S-Shaped model (DSSM). PSO algorithm will be usedto estimate the parameters of the well known SRGM.Detailed results and analysis are provided showingthe potential advantages of using PSO in solving thisproblem. Keywords: Particle Swarm Optimization, Soft-ware Reliability Growth Modeling, Software Testing. 1Introduction Software reliability is defined according to [21] as