A software test process stochastic control model based on CMM characterization

The state variable model of the software test process has shown to be a promising approach for the control of the system test phase of the software test process. However, the presence of unforeseen perturbations and noise in the data collection process impacts the accuracy of the predictions of the model and motivates the application of a stochastic control approach. This article explores a stochastic model construct upon the deterministic state variable model of the software test process. The Capability Maturity Model is used to identify levels of disturbance and noise that can be associated with the maturity level of processes. Simulation results are presented as an indication of the applicability of the stochastic state model. Copyright © 2004 John Wiley & Sons, Ltd.