Project cost overrun simulation in software product line development

  • Authors:

  • Makoto Nonaka;Liming Zhu;Muhammad Ali Babar;Mark Staples

  • Affiliations:

  • Faculty of Business Administration, Toyo University, Japan;National ICT Australia;Lero, University of Limerick, Ireland;National ICT Australia

  • Venue:
  • PROFES'07 Proceedings of the 8th international conference on Product-Focused Software Process Improvement
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The cost of a Software Product Line (SPL) development project sometimes exceeds the initially planned cost, because of requirements volatility and poor quality. In this paper, we propose a cost overrun simulation model for time-boxed SPL development. The model is an enhancement of a previous model, specifically now including: consideration of requirements volatility, consideration of unplanned work for defect correction during product projects, and nominal project cost overrun estimation. The model has been validated through stochastic simulations with fictional SPL project data, by comparing generated unplanned work effort to actual change effort, and by sensitivity analysis. The result shows that the proposed model has reasonable validity to estimate nominal project cost overruns and its variability. Analysis indicates that poor management of requirements and quality will almost double estimation error, for the studied simulation settings.