Abstract. In this paper we present a new test model written in SysML and an associated blackbox test suite for the Ceiling Speed Monitor (CSM) of the European Train Control System (ETCS). The model is publicly available and intended to serve as a novel benchmark for investigating new testing theories and comparing the capabilities of modelbased test automation tools. The CSM application inputs velocity values from a domain which could not be completely enumerated for test purposes with reasonable effort. We therefore apply a novel method for equivalence class testing that -despite the conceptually infinite cardinality of the input domains -is capable to produce finite test suites that are complete (i.e. sound and exhaustive) for a given fault model. In this paper, an overview of the model and the equivalence class testing strategy is given, and tool-based evaluation results are presented. For the technical details we refer to the published model and a technical report that is also available on the same website.
The application of test automation tools in a safety-critical context requires so-called tool qualification according to the applicable standards. The objective of this qualification is to justify that verification steps automated by the tool will not lead to faulty systems under test to be accepted as fit for purpose. In this paper we review the tool qualification requirements of the standards ISO 26262 (automotive domain) and the new RTCA DO-178C (avionic domain) and propose a general approach on how to qualify model-based testing tools according to these standards in an efficient and at the same time reliable way. Our approach relies on a lightweight error detection mechanism based on the idea of replaying test executions against the model. We further show how the error detection capabilities can be integrated into a convincing argument for tool qualification, going through the necessary verification activities step-by-step. We highlight the key steps for the RT-Tester Model-Based Test Generator, which is used in test campaigns in the automotive, railway and avionic domains. The approach avoids having to qualify several complex components present in model-based testing tools, such as code generators for test procedures and constraint solving algorithms for test data elaboration.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.