INVESTIGATING PERFORMANCE OF SOFTWARE PROJECTS EXECUTED FOR FORTUNE 500 COMPANIES

摘要

In this paper, Data Envelopment Analysis Variable Returns to Scale (DEA VRS) is applied to the data collected on 38 software development projects executed for Fortune 500 companies. The objective is to measure the productivity of software projects and to identify best practices in projects. In addition, this paper analyzes the root causes for productivity variations in the sample projects. Deviating from the prior research, this study segregates Total Efforts into Creation Efforts, which includes the Cost of Conformance, and Cost of Non-Conformance. The key findings are as follows. First, the results identify 11 projects that can be emulated for software process improvements and suggests that there is, on average, approximately 22 percent potential for productivity improvement. Second, the study validates that segregating total efforts has a positive impact on the best practice frontier.