Projects of the Experimental Software Engineering Group at the University of Maryland




As a first step in understanding where COTS-based development in the FDD stood, the study team analyzed the current data collection. Historically the SEL collects effort data. The SEL had made an attempt to gather data on COTS-related effort, but the level of detail was too general to allow understanding of the process. One indication that the SEL was not capturing useful data is the large amount of effort that fell into the "other" category.

Clearly, the quantitative information available was not sufficient for us to identify and understand the new issues that were arising in relation to the use of COTS packages in FDD projects. In order to gather more and richer information on this topic, the study team designed and conducted structured interviews, using three levels of interview guides at increasing levels of detail, with representatives from 12 projects. Topics covered included the process steps carried out, what problems were encountered with the use of COTS in development, and how the incorporation of COTS has changed the software development process.

Based on the interview data, we then described the COTS-based development process in the FDD, using a flow chart. Also based on interview data, we redesigned the way development effort data is collected in the FDD. These results were packaged through updated data collection forms and study briefs (short, Web-based descriptions of process and study results). The next level of understanding, then, is provided by the baseline process description and data from the new effort forms.

The COTS Study is transitioning into a second phase in which we will be focusing on three specific issues:

  • risk management in COTS-based development
  • developing cost models for COTS-based development
  • maintenance of COTS-based systems


Project Status

Active - three follow-on studies just beginning


There have been several contributions of the SEL COTS Study thus far:

  • a well-defined process characterization for package-based software development
  • lessons learned from COTS projects in the FDD
  • a new effort data collection form designed to capture the effort expended on various COTS-related activities
  • another data collection form that will help the SEL keep track of the various COTS products and tools that are being used in the FDD
  • the institution of Study Briefs in the SEL
  • further experience with the use of interviews in empirical studies

Top Level Page

Last updated: March 20, 1998 by Carolyn Seaman

Web Accessibility