Avispa: a tool for analyzing software process models. Issue 4 (11th February 2013)

• Record Type: Journal Article
• Authors: Hurtado Alegría, Julio A.; Bastarrica, María Cecilia; Bergel, Alexandre
• Other Names: Raffo David guestEditor.; Pfahl Dietmar guestEditor.; Zhang Li guestEditor.
• Is Part Of: Journal of software. Volume 26:Issue 4(2014:Apr.)
• Page Start: 434
• Page End: 450
• Publication Date: 2013-02-11
• Subjects: software visualization -- software process verification -- software process analysis; Software engineering -- Periodicals; Computer software -- Development -- Periodicals; Software maintenance -- Periodicals; 005.1
• DOI: 10.1002/smr.1578
• Languages: English
• ISSNs: 2047-7473

Abstract

ABSTRACT: Defining and formalizing the software development process is a common means for improving it. Software process modeling is often a challenging and expensive endeavor, because a well specified process may still include inefficiencies that are hardly detected before enacting it. Thus, assessing process quality is a relevant concern to improve several aspects such as conceptual integrity, correctness, usability, maintainability, and performance, among others. This paper describes Avispa, a graphical tool that allows analyzing the quality of SPEM 2.0 software processes models. Avispa identifies a series of error patterns and highlights them in different blueprints. A detailed description of the internals of Avispa is provided to show both its structure and its extensibility mechanisms. We also present an interactive mechanism to define new analysis scripts and to implement new patterns and blueprints. This paper illustrates the application of Avispa in an industrial case study where process engineers are assisted to analyze the quality of their process. Copyright © 2013 John Wiley & Sons, Ltd. Abstract : Software process assessment is determinant to ensure productive projects. We propose a set of visualizations to easily identify conceptual and structural anomalies in software processes. Our visualizations have been industrially validated.