Semi-quantitative Modeling for Managing Software Development Processes.
Abstract
Software process modeling has become an essential technique for managing software development processes. However, purely quantitative process modeling requires a detailed understanding and accurate measurement of software process, which relies on reliable and precise history data. This paper presents a semi-quantitative process modeling approach to model and manage software development processes. It allows for the existence of uncertainty and contingency during software development, and facilitates a manager's qualitative and quantitative estimates and assessments of process progress. We demonstrate its value and flexibility by developing semi-quantitative models of the test-and-fix process of incremental software development. Results conclude that the semi-quantitative process modeling approach can support process or project management activities, including estimating, planning, tracking and decision making throughout the software development cycle. 1. Introduction Software process modeling and analysis has been identified as being a vehicle for controlling development costs, duration, and achieving product quality. It has been widely accepted that improving the software process will also improve the software product quality, and increase project success rate. In recent years, software process modeling has been a major focus of software process improvement and is becoming one of the essential techniques for effectively managing software development processes. Nevertheless, purely quantitative process modeling requires very detailed understanding and accurate measurement of the software process, which relies on reliable and precise history data. Furthermore, in most cases these data are not readily available. 1530-0803/08 $25.00 © 2008 IEEE DOI 10.1109/ASWEC.2008.50 2. Background and approach 2.1. Software process modeling and simulation Osterweil identified two complementary types of software process research [4], which can be characterized as macro-process research, focused on 66 Authorized licensed use limited to: CITY UNIV OF HONG KONG. Downloaded on September 17,2026 at 02:49:10 UTC from IEEE Xplore. Restrictions apply. phenomenological observations