Using models for test generation and analysis

Abstract

Software testing will play a role in the development of software systems for some time to come. Although testing can account for 40 to 75 percent of the lifetime development and maintenance costs, the results summarized in this paper provide promising evidence that the use of test automation to support the manually intensive test generation and model-based analysis is feasible and practical. There is a great need to demonstrate and integrate new and advanced technologies. This paper describes an environment developed to validate the use of model-based translators on real-world applications. The environment integrates model-based development tools with a specification-based test vector generator and specification-based coverage analyzer.