Software maintenance process analysis using discrete-event simulation

Abstract

The competitive pressure and high costs of software maintenance force many IT organizations to carefully analyze and redesign their maintenance process. Identifying the changes that can remedy process deficiencies is not an easy task. Moreover, considerable risk is associated with implementing the proposed changes, since inadequate process design may cause a significant financial loss. The authors investigate the potential of simulation for analyzing and improving software maintenance processes. We present a simple, decision tree based model, suitable for software maintenance process simulation. The purpose of the model is to evaluate alternative process strategies and designs from a customer perspective. We apply the model to investigate deficiencies of the real software maintenance process and present a simulation experiment that examines an actual process design. Finally, we discuss the usability of simulation modeling for comparing and evaluating alternative process strategies in a real software process environment.