Abstract :
One major reason software development projects fail is that the development process is invisible. Managers tend to rely on meetings and reports to understand project status and make decisions, leading to mismanagement due to inaccurate or incomplete information. Software projects typically collect information during development using different tools and store it in repositories. We present a statistical process control (SPC) method of defining, collecting, and analyzing software metrics from software repositories for MDD process control and improvement (PCI). This method, which we call mining software repositories (MSR), can help us change the traditional, static, record-keeping use of software data repositories to a new, active use for predicting and planning various aspects of MDD projects
Keywords :
data mining; project management; software development management; software metrics; software process improvement; statistical process control; mining software repository; model-driven development; process control and improvement; software metrics; software project development; statistical process control; Costs; Guidelines; Phase measurement; Predictive models; Process control; Project management; Six sigma; Software measurement; Software tools; Telecommunication control; data mining; model-driven development; software metrics; statistical process control;
