Decision Sciences Journal 30(4) Index


Decision Sciences Journal
Volume 30, Number 4
Fall 1999

 

A Performance Measure for Software Reuse Projects

Marcus A. Rothenberger
School of Business Administration, University of Wisconsin – Milwaukee, PO Box 742, Milwaukee, WI 53201, e-mail: rothenb@uwm.edu

Kevin J. Dooley
Department of Management, Department of Industrial Engineering, Arizona State University, PO Box 874006, Tempe, AZ 85287-4006, e-mail: kevin.dooley@asu.edu

Abstract. The efficient development of software has become a critical factor to the success of many organizations. Significant competitive advantage can be realized by reducing the time to complete software development, while maintaining or improving quality levels. Project management activities, including the measurement of performance at the level of the project, can help the organization to both control projects and learn from experience. Traditionally, the productivity of a software development project has been measured as the complexity of code produced per unit of labor involved. An increasing number of organizations, however, are adopting the practice of software reuse, and under such conditions, such a traditional measure of productivity does not fully capture all the relevant elements of the reuse-driven development environment. This research develops a new performance measure that considers four major components of performance: productivity associated with development of new code, productivity associated with the reuse of previously written code, quality of the reuse decision, and relative value of reuse to the company. To illustrate the feasibility of the measure, a real-world productivity assessment is presented for an example company, and related implementation issues are discussed.

Subject Areas: Computer Science, Heuristics, Management Information Systems, Parameter Estimation, Performance Appraisal, Productivity, Quality, Statistics, and Systems Development Methodologies.

back to 30(4) Index

DSI Home Page