Software Testing by Statistical Methods: Preliminary Success Estimates forApproaches Based on Binomial Models, Coverage Designs, Mutation Testing, and Usage Models

摘要

Software conformance testing is the process of determining the correctness of animplementation built to the requirements of a functional specification. Exhaustive conformance testing of software is not practical because variable input values and variable sequencing of inputs result in too many possible combinations to test. This paper addresses alternatives for exhaustive testing based on statistical methods, including experimental designs of tests, statistical selection of representative test cases, quantification of test results, and provision of statistical levels of confidence or probability that a program implements its functional specification correctly. The goal of this work is to ensure software quality and to develop methods for software conformance testing based on known statistical techniques, including multivariable analysis, design of experiments, coverage designs, usage models, and optimization techniques, as well as to provide quantitative measures of quality, reliability, and conformance to specifications, including statistical measures and confidence levels.