Model Based System Development/Move or Clean UP/Model Based Testing

Model Based Testing

Quality of software is an issue of increasing importance and growing concern. Systematic testing of software plays an important role in the quest for improved quality. Testing, however, turns out to be expensive and time consuming. One source of problems is the often imprecise and ambiguous nature of specifications, so that a good basis for testing is missing. Another reason is the usually manual and laborious testing process without effective automation, so that testing is error-prone and consumes many resources.

Current research efforts in model based testing aim at contributing to solutions for these testing problems. The starting point for model based testing is a formal description, or model, of the system under test. On the one hand, such a model serves as a precise and complete specification, and, consequently, is a good basis for testing. On the other hand, such models can be automatically processed by tools, which allows to effectively automate the testing process. In particular, algorithms have been developed to generate tests from such formal models. Tools, implementing these algorithms, lead to automatic, faster and less error-prone test generation: millions of test events can be automatically generated, and `on-the-fly' executed, and analysed.

Model based testing is an important aspect of the project TANGRAM, which is a joint research and development effort of a couple of universities and companies, coordinated by the Embedded Systems Institute, in which ASML is the core industrial partner.

Keywords: software testing, model-based testing, specification-based testing, formal methods, test automation, test generation.