Smart equipment as a cyber-physical system can be envisioned to support
owner/operators with model-based assistance systems. In fluctuating process scenarios, which are expected within modular plants, frequent adaptation of the required simulation models to changing conditions will be necessary. Therefore, automated functional quality assessment for simulation models would provide a significant added value for the owner/operator. Currently, quality assessment of simulation models is dominated by the verification and validation methods developed by Sargent and co-authors (Sargent and Balci, 2017). Although these methods provide a wide spectrum of assessment methods, they base on a rather narrow view on quality of simulation models. In addition, quality and quality advancement are typically not measured continuously. To overcome these obstacles, the authors discuss the applicability of central quality assurance methods from the field of software development to simulation model assessment. These are the principle of test-driven development and the application of quality models to measure quality and quality advancement. A framework and a workflow for test-driven development of simulation models are proposed and evaluated in a case study on a soft sensor for a fermentation process. It was found, that in principle, the framework provides transparent and continuous quality assessment in a semi-automated manner and lays the foundation for guidance strategies during model development.

Mädler, J., Viedt, I., & Urbas, L. (2021). Applying quality assurance concepts from software development to simulation model assessment in smart equipment. In Computer Aided Chemical Engineering (Vol. 50, pp. 813-818). Elsevier.