| Software testing is the process of executing software to experimentally validate its correctness with respect to its specification. Failures are deviations from the specification and are determined by comparing observed vs. intended behavior of the system. Software testing is inherently an unbounded process because it will never be known with total confidence whether the software under test contains no latent faults or will never fail. Thus any testing process is a form of sampling. The tester uses some knowledge of the system to select a finite subset of tests to execute. Too often, this selection process is ad hoc or guided by structural coverage of the source code.; Software systems have evolved into some of the most complex artifacts ever created by humans. However, the quality of software has been a main concern since the inception of computer software. This complexity has exacerbated the problems of testing software. Years ago, it was shown that testing a simple program to add only two integer inputs of 32-bits (yielding 2∧ 64 distinct test cases) would take hundreds of years, even if tests were performed at a rate of thousands per second. To be able to produce high quality software, software developers and software testers alike need continuous improvements in development and testing methodologies. These improvements should result in better coverage of the input domain and decreased testing costs. Testers are increasingly expected to verify software quality in less time and with fewer resources. In such an environment, solid, repeatable, and practical testing methods and automation are a must.; In this research we focus on an approach of generating efficient test cases based on the special properties of Design of Experiments (DOE) and developing a logistic regression model of predicting test case outcomes. This approach, in turn, would provide the software tester with a model that reduces the number of test cases, predicts test case outcomes, reduces cost, and allows better forecast of release readiness. The outcome of this research is validated through two cases studies. |
