Statistical analysis of deviation of actual cost from estimated cost usingactual project data

This paper analyzes an association of a deviation of the actual cost (measu red by person-month) from the estimated cost with the quality and the produ ctivity of software development projects. Although the obtained results the mselves may not be new from the academic point of view, they could provide motivation for developers to join process improvement activities in a softw are company and thus become a driving force for promoting the process impro vement. We show that if a project is performed faithfully under a well-organized pr oject plan (i.e. the plan is first constructed according to the standards o f good writing, and then a project is managed and controlled to meet the pl an), the deviation of the actual cost from the estimated one becomes small. Next we show statistically that projects with small deviation of the cost estimate tend to achieve high quality of final products and high productivi ty of development teams. In this analysis, the actual project data on 37 pr ojects at a certain company are extensively applied. (C) 2000 Elsevier Scie nce B.V. All rights reserved.