Validating Production Test Scenarios with Cyber-Physical System Design Models

Abstract

Information systems are key facilitators for production planning, configuration, and monitoring in increasingly flexible Cyber-Physical Production Systems (CPPSs), such as automated welding lines for joining car parts. The guideline VDI 2206 provides a V-model to organize the engineering disciplines. It emphasizes the importance of validation in CPPS engineering but provides limited guidance on the up-front specification and validation of test scenarios. This paper introduces the Production Test Scenario Validation (PTSV) approach to validate to what extent a multi-view CPPS design model and associated data sources from production engineering and operation represent the concepts used in test scenarios. The PTSV (i) elicits concepts in pre- and post-conditions of production test scenarios, (ii) maps these concepts to a multi-view CPPS design model, such as a production asset network, and (iii) identifies mismatches between production test scenarios and CPPS design models, such as missing model elements or data sources required for validation. We evaluated the effectiveness and efficiency of the approach in a feasibility study on a real-world welding line for joining car parts. The results indicate that the PTSV is feasible and effective for improving the capability to measure and analyze outcomes of test scenarios in CPPS engineering according to the VDI 2206.