Optimized Software Quality Assurance Model for Testing Scientific Software

Software projects in R&D organizations differ in their quality assurance process compared to other production and business organizations. Major hard constraints are accuracy and precision. Estimated number of defects the product is likely to contain at release shall be as minimum as possible. Various models for assessing the quality of the software are developed and are in use. The most widely used models are McCall‟s, ISO 9000, CMM and COCOMO. These models sometimes become laborious during testing. Hence it is necessary to optimize the methodology of software quality assurance. So that it becomes robust, fast and economic. Based on the requirements an effort is made in this project to develop “Optimal Software Quality Assurance Model for Testing Scientific Software” which shall produce reliable and robust software engineering model to meet the requirements of IEEE 12207. Here through several procedures such as clustering of requirements, mapping, and so on, are used in order to find out the Defect Density of software and to predict its reliability and quality. The optimized model can overcome the process limitations of traditionally applied models and to provide an efficient way to assess the quality and other factors pertaining to scientific software systems. The optimized model is validated through comparing previous test results with results obtained from applying this model and the model was found working as per the requirements.