7. Conclusion

In this chapter we presented a practical approach to software defect prediction, which helps assure the delivery of high quality software. An innovative cloud-based analytics tool, BRACE, was introduced which automates the entire process of data extraction, pre-processing, core processing, and post-processing, combined with a user interface. It no longer relies on the use of a spreadsheet and generates prediction in real-time, which can be shared with any members of a project. SRGM is the core analytics engine which implements technical breakthroughs in this area. It provides a robust, consistent, flexible, fast, statistically sound approach to defect prediction for any defect data sets without human intervention. The enhanced version of SRGM incorporates feature arrival data to provide defect prediction throughout the lifecycle of each release with much improved accuracy. We also demonstrated the method for predicting customer defects and software availability during the operation phase, which should be the basis for software quality assurance. We demonstrated the effectiveness of the approach using data sets taken from telecom development projects, varying from traditional development to DevOps CI/CD with full agile development. This approach can be easily applied to any software development projects.
