QWE2002 Paper 6T Dr. Stacy Prowell (The University of Tennessee) Tool Support for Model Based Statistical Testing

Key Points

• Model-based statistical testing is both efficient and cost-effective.
• Model-based testing supports the use of test automation tools.
• Significant tool support for model-based testing has been developed.

Presentation Abstract

This talk gives an overview of statistical testing based on a usage model, focusing on opportunities for automation in the process. In particular, an overview of the Java Usage Model Builder Library (JUMBL), a Java class library to support statistical testing, is given. The JUMBL supports the development and analysis of usage models, generation and transformation of tests, and analysis of test results.

About the Author

Dr. Prowell is a Research Associate Professor in the Department of Computer Science at the University of Tennessee where his work focuses of rigorous software specification and statistical testing. Prior to joining the Univerity of Tennessee, Dr. Prowell worked for Q-Labs on industrial application of process improvement and statistical testing techniques.