STUDIES IN SIMULATION AND GAMING
Online ISSN : 2434-0472
Print ISSN : 1345-1499
Peer-Reviewed Paper
An Evaluation Study of Requirements Review Techniques Based on a Crossover Trial
Koichi Tanizaki Shuang ChangHiroshi Deguchi
Author information
JOURNAL FREE ACCESS

2020 Volume 30 Issue 2 Pages 95-109

Details
Abstract

Creating requirements specifications is essential in product development projects. It is important to detect defects of requirements specifications by reviewing them in a phase of requirements analysis because it can lead big cost to fix the defects if they are found in post-process. Reviewing of requirements specifications is a form of manual simulation and reviewers should consider various boundary conditions to simulate the behavior of products and to detect defects. Considering boundary conditions that are not written in requirements specifications is difficult with conventional review techniques such as Checklist-based reviewing and Scenario-based reviewing. To solve this problem, a review technique based on adverse conditions analysis (ACA) was proposed. ACA incorporates modeling of boundary conditions. According to previous research, ACA is effective to review requirements specifications. However, discussion about the difference between ACA and conventional review techniques was insufficient. Verification of the effectiveness of it was also not enough. Therefore, we clarified its characteristics and confirmed effectiveness through experiment based on a crossover trial that test engineers conduct reviewing of requirements specifications using ACA and the conventional review technique. It was confirmed that many important defects can be detected by reviewing requirements specifications using ACA and that it was cost-effective. We also discussed the application of gaming to ACA.

Content from these authors
© 2020 Japan Association of Simulation & Gaming
Previous article Next article
feedback
Top