Optimizing Systems and Software Project Management

Systems and software projects traditionally experienced some degree of project failure. In this context, failure is defined as a project that demonstrated a failure to match, within a reasonable tolerance, the expected outcome. Successful software projects demonstrate good project management methods incorporating modern processes and effective utilization of
metrics. This article extends the application of control theory methods to project management decisions to include optimization of systems trade-offs to improve project performance.

Applying the feedback control theory to the management decision process improve performance of the project process. Usage of parametric models is well accepted for software project estimation and planning. Maintaining the models at a high fidelity level for the duration of a project provides high confidence estimates of expected cost to complete and optimized selection of corrective actions. Updated models calibrated to the actual performance provide a confidence level estimate of future performance. The updated models generate stoplight charts and a set of information similar to EVM outputs but based on project estimates from a calibrated model. Cost and effort parametric estimating models provide a powerful aid to project management, yielding “objective information” for systems trade-offs, project management decisions and controlling project performance.