Performance measures for supporting project manager decisions

Software measurement is a crucial technology along the software development process, and notably for project management. Decision-making in project management requires in fact, accurate estimations and the support of metrics to monitor and control the varying factors affecting the development process. Managers must make reliable schedule predictions and optimize personnel utilization. They therefore must be able to dynamically evaluate whether the resources assigned to a job are sufficient and whether the organization structure is adequate to meet the scheduled deadlines. To support managers in these tasks, we propose a method called Propean (for PROject PErformance ANalysis) based on the combination of classical performance engineering techniques and the Unified Modelling Language (UML). The combined application of these disciplines guarantees on one hand, a validated approach for modelling and estimating the 'Quality of Service' parameters when they come to the development process. Performance engineering techniques are in fact based on rigorousmathematical formalisms such as queueing networks, stochastic Petri nets and Markov models. On the other hand, the adoption of UML provides managers with an easy-to-use front-end representation of the process closer to the design notations they routinely employ. To illustrate Propean application, in this article, we model the case of a manager who must decide a release date for a product undergoing the testing phase.