Lessons learned from automated analysis of industrial UML class models (an experience report)

• Published in: Lecture notes in computer science pp. 324 - 338
• Main Authors: Cheng, Betty H. C., Stephenson, Ryan, Berenbach, Brian
• Format: Conference Proceeding Book Chapter
• Language: English
• Published: Berlin, Heidelberg Springer-Verlag 01.01.2005; Springer Berlin Heidelberg; Springer
• Edition: 1ère éd
• Series: ACM Conferences
• Subjects: Applied sciences; Automate Analysis; Computer science; control theory; systems; Design Error; Design Pattern; Exact sciences and technology; Lesson Learn; Software; Software and its engineering; Software and its engineering > Software creation and management; Software and its engineering > Software creation and management > Software development process management; Software engineering; Software Metrics; Data analysis; Software development; Process planning; Reverse engineering; Design pattern; Reuse; Model driven architecture; Modeling; Design process; Software tool; Object oriented; Software quality; Unified modelling language; Metric; Metamodel
• ISBN: 9783540290100; 3540290109
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/11557432_24

Automated analysis of object-oriented design models can provide insight into the quality of a given software design. Data obtained from automated analysis, however, is often too complex to be easily understood by a designer. This paper examines the use of an automated analysis tool on industrial software UML class models, where one set of models was created as part of the design process and the other was obtained from reverse engineering code. The analysis was performed by DesignAdvisor, a tool developed by Siemens Corporate Research, that supports metrics-based analysis and detection of design guideline violations. The paper describes the lessons learned from using the automated analysis techniques to assess the quality of these models. We also assess the impact of design pattern use in the overall quality of the models. Based on our lessons learned, identify design guidelines that would minimize the occurrence of these errors.