Applying DPPI: A Defect Causal Analysis Approach Using Bayesian Networks

Defect causal analysis (DCA) provides a means for product-focused software process improvement. A DCA approach, called DPPI (Defect Prevention-based Process Improvement), was assembled based on DCA guidance obtained from systematic reviews and on feedback gathered from experts in the field. According to the systematic reviews, and to our knowledge, DPPI represents the only approach that integrates cause-effect learning mechanisms (by using Bayesian networks) into DCA meetings. In this paper we extend the knowledge regarding the feasibility of using DPPI by the software industry, by describing the experience of applying it end-to-end to a real Web-based software project and providing additional industrial usage considerations. Building and using Bayesian networks in the context of DCA showed promising preliminary results and revealed interesting possibilities.

[1]  George Eckes,et al.  The Six Sigma Revolution: How General Electric and Others Turned Process Into Profits , 2000 .

[2]  Forrest Shull,et al.  Developing techniques for using software documents: a series of empirical studies , 1998 .

[3]  David N. Card,et al.  Defect Analysis: Basic Techniques for Management and Learning , 2005, Adv. Comput..

[4]  William A. Florac,et al.  Measuring the Software Process: Statistical Process Control for Software Process Improvement , 1999 .

[5]  P. Jalote,et al.  Using defect analysis feedback for improving quality and productivity in iterative software development , 2005, 2005 International Conference on Information and Communication Technology.

[6]  Tanja Toroi,et al.  Difficulties in Establishing a Defect Management Process: A Case Study , 2006, PROFES.

[7]  Robert B. Grady,et al.  Software Failure Analysis for High-Return Process Improvement Decisions , 1996 .

[8]  Min Xie,et al.  A statistical method for controlling software defect detection process , 1999 .

[9]  Guilherme Horta Travassos,et al.  A computational framework for supporting software inspections , 2004 .

[10]  Stuart Anderson,et al.  Requirements Evolution from Process to Product Oriented Management , 2001, PROFES.

[11]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[12]  Leon G. Higley,et al.  Forensic Entomology: An Introduction , 2009 .

[13]  Guilherme Horta Travassos,et al.  Towards a Defect Prevention Based Process Improvement Approach , 2008, 2008 34th Euromicro Conference Software Engineering and Advanced Applications.

[14]  Joseph G. Voelkel,et al.  Guide to Quality Control , 1982 .

[15]  Lars Lundberg,et al.  Company-Wide Implementation of Metrics for Early Software Fault Detection , 2007, 29th International Conference on Software Engineering (ICSE'07).

[16]  D.N. Card.  Defect-causal analysis drives down error rates , 1993, IEEE Software.

[17]  Carole L. Jones.  A Process-Integrated Approach to Defect Prevention , 1985, IBM Syst. J..

[18]  Robert G. Mays,et al.  Experiences with Defect Prevention , 1990, IBM Syst. J..

[19]  Pearl Brereton,et al.  Lessons from applying the systematic literature review process within the software engineering domain , 2007, J. Syst. Softw..

[20]  Dewayne E. Perry,et al.  Classification and evaluation of defects in a project retrospective , 2002, J. Syst. Softw..