This presentation was made at the NAFEMS Americas Seminar "Model-Based Engineering: What is it & How Will It Impact Engineering Simulation" held on the 1st of October 2019 in Columbus Ohio

Resource Abstract

With the ever-increasing speed of technology advancements in development of cyberphysical systems, there is a corresponding struggle to grasp system complexity and to establish an efficient development process.



The different verification and validation (V&V) phases are becoming an increasingly vital part in today’s engineering processes having to address stricter safety and environmental regulations. A model-based design methodology is widely adopted for verification and validation, but its true potential is being limited as simulation is used within an ecosystem of often disconnected engineering activities.



Due to the lack of a platform providing traceability and reliability support for collaborative engineering, test results often reflect outdated development status or regression testing after requirement changes cannot be automated. Consequently, engineers are discovering bugs only at final integration or system testing, even though approximately 60% of them are already introduced at the requirements phase.



OEMs and suppliers require integrated platforms that can describe and orchestrate the processes between various engineering activities and stakeholders, including the management of large amounts of requirements or test cases and the connection to simulation for running validation and verification scenarios and cross-domain integration activities.



In this presentation, we introduce such an innovation platform that allows companies to sketch out, orchestrate and optimize the different processes currently followed in their development cycles. To achieve this a technology concept called Model Identity Cards (MIC) is introduced. The MIC framework is built on three main pillars, of which each will be given an introduction and more technical explanation.



The first pillar focuses on providing traceability and reliability to the engineering community. This can be ensured by creating bidirectional links between the requirements, test cases and all simulation artefacts created during the implementation and testing phases. A formalized information model implemented in the backend gives an overview of the possible links that can be created. This information model then also creates the basis for a plug-and-play platform where artefact types can be added and removed based on a company’s best practices. On each of these artefact types, a formalized workflow is then imposed to create a reliable business process that is adopted company-wide. This is for example important in context of certification.



The second pillar of the platform is an architecture model of the simulation components. This model allows the engineering team to describe and monitor how the different components will integrate with one another. The architecture model furthermore streamlines the communication to successive development teams, who can then implement executable simulation components in their simulation platform of interest. Successively, by being able to retrieve up-to-date development and test information on related simulation components, the architecture model forms the basis for the automation of regression tests.



The third pillar is the specification and execution of the V&V platform. Based on the available tooling, methods, and IT infrastructure, the MIC connects to an automated server-based test framework. Built on a continuous integration solution, this framework sets up a co-simulation environment based on the information received from the MIC. It then executes this environment to validate the executable simulation models. Results from these tests again complete the Model Identity Cards, facilitating re-use of models, documentation generation and clear results reporting between different development teams.



To summarize, the presentation will introduce the concept of Model Identity Cards to the audience. The talk will focus on the supported processes, will give insights in the back end metamodels and will give an outlook on how we plan to further extend capabilities. Finally we’ll discuss how adopting Model Identity Cards for simulation could streamline collaboration, reduce design iterations and reduce manual testing effort by automating verification and validation activities.