Determining an Optimal Time Interval for Testing and Debugging Software

摘要

In this paper, we describe an approach for addressing the important problem ofhow long to test and debug software before it is released. Our approach is based on the principles of decision making under uncertainty, and involves a maximization of expected utility. We suggest two plausible forms for the utility function, one based on costs alone and the other involving the realized reliability of the software. Using results from the literature on probabilistic models for software failure, we outline, for the case of single state testing, the ensuing optimization problem which can be addressed using numerical techniques. The sensitivity of our results to the various input parameters is