Abhinav Sharma, HCL Technologies, UK

A simple model used successfully for estimating and tracking software defects to predict launch readiness of software in a complex product is described in this paper. The model is based on tracking the number of defects estimated to be found, actually found and resolved to measure the quality of the product. Defect estimates can also help identify quality and process issues in the development and testing phases. The defect estimation tracking method described here covers the whole project and is split into the three phases Initial Defect Estimates (based on historical data), Interim Revised Estimates (based on actual performance of the project) and Final Defect Tracking (based on testing still to do). The method is based on existing development processes of the team so is easier to implement and has been successfully applied in several projects.

Software Reliability Growth Model, Defect Estimation, Software Quality Tracking, Schedule Prediction