An Enhanced Pairwise Search Approach for Generating Optimum Number of Test Data and Reduce Execution Time

Abstract

In recent days testing considers the most important task for building software that is free from error. Since the resources and time is limited to produce software, hence, it is not possible of performing exhaustive tests (i.e. to test all possible combinations of input data.) An alternative to get ride from this type exhaustive numbers and as well to reduce cost, an approach called Pairwise (2 way) test data generation approach will be effective. Most of the software faults in pairwise approach caused by unusual combination of input data.  Hence, the demand for the optimization of number of generated test-cases and reducing the execution time is growing in software industries. This paper proposes an enhancement in pairwise search approach which generates optimum number of input values for testing purposes.  In this approach it searches the most coverable pairs by pairing parameters and adopts one-test-at-a-time strategy for constructing a final test-suite.  Compared to other existing strategies, Our proposed approach is effective in terms of number of generated test cases and of execution time.

Keywords:, Software testing, Pairwise testing, Combinatorial interaction testing, Test case generation.