A Requirement Based Approach to Test Case Prioritization for Regression Testing

Testing is the strategy wherein a program is executed fully intent on tracking down bugs. Whenever modification is done it is verified again to ensure that there are no new flaws have emerged in the previously tested code and that it continues to function effectively. This kind of testing is called regression testing. Original program may have many test cases and executing the whole test suite may increase the cost of testing. Regression test selection, test suite minimization and the test case prioritization are three strategies of regression testing which lower cost of regression testing. In test case selection and minimization techniques, only few test cases are chosen and the leftover test cases are squandered. The risk is that useful test cases with respect to fault detection may be discarded while in prioritization the software tester prioritizes the test and test cases that are more suitable to detect errors early are run first. There is no risk of important test case elimination in this technique as compared to other techniques. Therefore, test case prioritization is more effective and time saving. In test case prioritization technique important test cases are placed first in the prioritization list. Maximize early fault detection is goal of prioritization. In prioritization, technique test cases are ranked according to some coverage criteria. Existing techniques use total coverage strategy for prioritization of test cases by considering the total number of requirements covered. There may be more than one test case that covers the same requirements. Therefore, there is a possibility to repeat the test cases that cover same requirements, which is not desirable. In this research, we offer a new test case prioritizing technique based on prioritized requirements. This algorithm uses the "Additional" coverage strategy. Test case that covers the most requirements and has the highest weight is given high priority by our algorithm. Our suggested prioritization algorithm outperforms than existing techniques as it gives earlier coverage of requirements.