A case study in defect measurement and root cause analysis in a turkish software organization

Atagoren C., Chouseinoglou O.

Studies in Computational Intelligence, vol.496, pp.55-72, 2014 (Peer-Reviewed Journal) identifier

  • Publication Type: Article / Abstract
  • Volume: 496
  • Publication Date: 2014
  • Doi Number: 10.1007/978-3-319-00948-3_4
  • Journal Name: Studies in Computational Intelligence
  • Journal Indexes: Science Citation Index Expanded, Scopus, zbMATH
  • Page Numbers: pp.55-72


In software projects, final products aim to meet customer needs and concurrently to have the least number of defects. Defect identification and removal processes offer valuable insights regarding all stages of software development. Therefore, defects are recorded during the software development process with the intentions of not only fixing them before the product is delivered to the customer, but also accumulating data that can be researched upon. That data can later be used for software process improvement. One of the techniques for analyzing defects is the root cause analysis (RCA). A case study is conducted in one of the leading, medium sized software companies of Turkey by utilizing the RCA method. The collected defect data has been analyzed with Pareto charts and the root causes for outstanding defect categories have been identified with the use of fishbone diagrams and expert grading, demonstrating that these techniques can be effectively used in RCA. The main root causes of the investigated defect items have been identified as lack of knowledge and extenuation of the undertaken task, and corrective actions have been proposed to upper management. The case study is formulated in a way to provide a basis for software development organizations that aim to conduct defect analysis and obtain meaningful results. All stages of the research and the case study are explained in detail and the efforts spent are given. © 2014 Springer International Publishing Switzerland.