Software Test Case Selection Based on Comparative Cost and Risk-Based Value
The testing challenge is being able to select an efficient and effective set of test cases within the limited resources available for testing. The term “efficient” refers to the realization that the number of test cases is inherently limited by the testing resources available. The term “effective” refers to the ability of the selected test cases to find code defects that exist. Mark Gillenson addressed this problem by interviewing a set of testing professionals in a corporate partner and applying qualitative research techniques to reach their conclusions. This work was conducted in two stages. The first stage was to develop a comparative cost model of test cases. The second stage was to incorporate the cost model results into a broader model that focused on the comparative risk-based value of the test cases. The models were realized in the form of mathematical functions that could be used to compare the cost and value of potential test cases. The results were evaluated using a Monte-Carlo simulation.
Dr. Mark L. Gillenson is the Director of the Systems Testing Excellence Program (STEP) in the FedEx Institute of Technology at the University of Memphis, where he is University Research Professor in the Business Information and Technology Department of the Fogelman College of Business and Economics. STEP, founded in 2006 as a collaboration between the University of Memphis and the FedEx Corporation, has conducted numerous research projects in software testing and has trained thousands of software testing professionals in industry, the US Department of Defense, and the US Department of Homeland Security. Dr. Gillenson worked for the IBM Corp. for 15 years full-time and undergraduate summers dating back to 1965. Dr. Gillenson holds a PhD in Computer and Information Science from the Ohio State University.