A Process-Oriented System Dynamics Model for Software Development Project Prediction

Abstract

Project management is a typically experience centralized and process-driven activity. Various critical factors with interaction effect would decide project management success leads to complexity project management prediction problem. Project managers indeed require some assistance to reduce the uncertainty. To improve the predicament of project managers, this research built a process-oriented dynamics model that provides project managers a comprehensive systemically perspective to improve project prediction accurate rate, including: the project complexity prediction, the project staff allocation and the cost-revenue evaluation. As the experiment results demonstrated, the proposed model was successfully applied to simulate 41 real projects from an IT consultancy in Taiwan. The accurate rate of project complexity prediction is average 90%. Additionally, the proposed model enable project managers to simulate project staff allocation, including: PM, SA, SD and project programmer. Finally, the cost-revenue evaluation provide above 65% projects could achieve positive profit. These simulation results could successfully provide information for PM to reduce project management uncertainty.