Factors Influencing Requirements Traceability Practice

Author(s): Ramesh, B.
Venue: Communications of the ACM
Date: December 1998

Type of Experiement: Survey/Multi-Case Study
Sample Size: 138
Class/Experience Level: Professional
Data Collection Method: Observation, Survey, Project Artifact(s)


Link: http://portal.acm.org/citation.cfm?id=290133.290147&coll=ACM&dl=ACM&CFID...

This paper presents data gathered from four years worth of empirical studies. The studies include 28 focus groups in 26 organizations representing a wide variety of industries (defense, government, aerospace, hardware development, pharmaceuticals, utility, system integration, electronics, and telecommunications). The focus groups were followed by structured interviews. The projects involved range in size from $15 million to several billion, several thousand lines of code to 1 million lines of code, and 600 to 10,000 requirements.

The results show that participants can be categorized as "low-end" traceability users and "high-end" traceability users. Low-end users typically have 0-2 years of experience with traceability and work on projects with about 1,000 requirements. High-end users typically have 4-10 years of experience with traceability and work on projects with about 10,000 requirements.

Low-end and high-end users also differ in their attitudes toward traceability. Low-end users view traceability merely as a mandate, whereas high-end users view traceability as an integral part of quality systems engineering. Low-end users tend to not capture process-related traceability information such as rationale behind decisions. High-end users see traceability as a means to understand and improve their processes, which can give them a competitive advantage to help win future contracts or save money by reusing traceability information in the future.