Framework for Flexible, Adaptive Support of Test Management by Means of Software Agents
| Autor: | Franziska Maurer, Claudius V. Jordan, Sven Lowenberg, Julien Provost |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Control and Optimization
Test management business.industry Computer science Mechanical Engineering Biomedical Engineering System testing Context (language use) ddc Computer Science Applications Test (assessment) Human-Computer Interaction Test script Test case Artificial Intelligence Control and Systems Engineering Software agent Resource allocation Computer Vision and Pattern Recognition Software engineering business |
| Zdroj: | IEEE Robotics and Automation Letters. 4:2754-2761 |
| ISSN: | 2377-3774 |
| DOI: | 10.1109/lra.2019.2918486 |
| Popis: | In the context of system testing on hardware-in-theloop test benches, performing functional tests on electrical control units comes with various challenges. Especially planning the execution of test cases for a test cycle and nightly batch runs is cumbersome. In order to support the test managers, this letter proposes an adaptable and flexible test management assistance system. Therefore, a set of requirements for such an assistance system has been identified and proposed solutions from the literature have been recalled. Out of those, software agents appear to be a suitable method for building up the framework of the assistance system. The framework proposed in this letter is divided into two phases: static prioritization of test scripts and dynamic execution ordering. The agents of the former phase aggregate relevant available data from various sources to calculate a test script priority. In the latter, the agents negotiate to solve the resource allocation problem for test scripts on test benches. In addition, test logs are taken into account at runtime for reprioritization. Finally, an industrial use case from the automotive industry is taken as an example to discuss the applicability and current obstacles on the way to application and qualitative evaluation of the expected benefits are discussed. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|