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

Simulation is key in driving data-driven decision-making throughout the product design and development process, creating a virtual environment where engineers can optimize and make design trade-offs for cost and function. While simulation practices are well established in the design and development phase, physics-based models are rarely leveraged after a product is released. These physics-based simulation models provide insights into the operational conditions of the product, which cannot be captured by sensors. These models can enable “what if analysis” and diagnosis of problems, to aid the decision-making process well into the operations and sustainment cycle. At the same time, high-investment industries, like aerospace and defense, support programs that are planned for use over decades, in a highly evolving technology environment. These programs need to be designed not just for a successful roll-out but for continuous development and upgrades over the product lifecycle. Modularity, therefore, needs to be considered up-front as part of any digital engineering methodology, starting with the architecture, not only as part of a solid MBE strategy, but to ensure the safety and robustness of the system. The architecture of these high-investment, safety-critical systems needs to be validated against the threat of cyberattacks. It is critical, at the early architecture stage, to identify system vulnerabilities that can be exploited to execute attacks, so the appropriate measures can be taken. Embedded software, which defines the actions of systems, can help provide scalable, fully-integrated avionics software, to meet strict standards as well as development timelines, in a modular environment. For such an environment, it is essential that the digital design model database is maintained in an open format that seamlessly accesses the existing model data and can add new model data based on the latest practices and standards. Requirements-driven engineering is on the horizon and the MBE paradigm will create the need for modularity of simulations for engineers to "plug and play" from existing libraries of simulations. Model requirements, therefore, need to be defined up-front, in order for engineers to leverage insights from physics-based models which evolve from design to deployment and through operations, spanning from embedded software to Digital Twins.