Impact of Requirements Quality on Project Success or Failure
We are interested in the relationship between the quality of the requirements specifications for software projects and the subsequent outcome of the projects. To examine this relationship, we investigated 32 projects started and completed between 2003 and 2005 by the software development division of a large company in Tokyo. The company has collected reliable data on requirements specification quality, as evaluated by software quality assurance teams, and overall project performance data relating to cost and time overruns. The data for requirements specification quality were first converted into a multiple-dimensional space, with each dimension corresponding to an item of the recommended structure for software requirements specifications (SRS) defined in IEEE Std. 830-1998. We applied various statistical analysis methods to the SRS quality data and project outcomes.
The results showed some interesting relationships between the quality of the requirements and the success or failure of projects; for example, (1) a relatively small set of SRS items had a strong impact on whether a project succeeded or failed; (2) descriptions of SRS in normal projects tended to be balanced; (3) SRS descriptions in Section 1, which were expected to include the purpose, overview and general context for SRS, were comprehensive for normal projects but inadequate for projects that finished with overruns; and (4) when the descriptions of SRS in Section 1 were inadequate, while those of the expected functions and product perspective were comprehensive, the project tended to end up with cost overruns.
KeywordsRequirements quality project success statistical analysis
Unable to display preview. Download preview PDF.
- 1.Berry, D., Damian, D., Finkelstein, A., Gause, D., Hall, R., Simmons, E., Wassyng, A.: To do or not to do: If the requirements engineering payoff is so good, why aren’t more companies doing it? In: Proc. 13th International Requirements Engineering Conference (RE 2005), p. 447. IEEE, Los Alamitos (2005)CrossRefGoogle Scholar
- 2.Standish Group International: The chaos report (1994), http://www.standishgroup.com/sample_research/chaos_1994_1.php
- 3.Glass, R.L.: The standish report: Does it really describe a software crisis? Communications of the ACM 49, 15–16 (2006)Google Scholar
- 5.IEEE: Recommended practice for software requirements specifications. Technical report, IEEE, IEEE Std 830-1998 (1998)Google Scholar
- 7.The R Foundation for Statistical Computing: The R project for statistical computing, http://www.r-project.org/index.html
- 9.Wohlwend, H., Rosenbaum, S.: Software improvements in an international company. In: 15th International Conference on Software Engineering (ICSE 1993), Baltimore, MD, USA (1993)Google Scholar
- 11.Verner, J., Cox, K., Bleistein, S., Cerpa, N.: Requirements engineering and software project success: An industrial survey in Australia and the U.S. Australian Journal of Information Systems 13, 225–238 (2005)Google Scholar
- 12.Emam, K.E., Madhavji, N.H.: Measuring the success of requirements engineering processes. In: Second IEEE International Symposium on Requirements Engineering, pp. 204–211 (1995)Google Scholar