“Benchmark” Issue: “The Need for Speed” and – inevitable in these times – “COVID-19”
“Speed” was the originally intended theme for issue No. 2 of the magazine “Benchmark”, the international magazine for engineering designers and analysts from NAFEMS, in 2020. NAFEMS is the international association for the engineering modelling, analysis and simulation community, a not-for-profit organization established in 1983. Dr.-Ing. Volker Gravemeier, chief executive officer of AdCo EngineeringGW, is an active member of the NAFEMS Multiphysics Working Group (NAFEMS MWG). However, as David Quinn points out in his editorial, it “turns out that life can change quicker than anything else”. In this blog post, we will have a closer look at two articles of this issue.
Peter Langsten and Marc Halpern identify digital strategies for “Overcoming COVID-19 Obstacles” in their contribution. They conclude that “the COVID-19 pandemic will accelerate many important long-delayed trends such as digitalization, integrated flow of product information, and collaborative applications. The post COVID-19 engineering environment will pose new demands… The new reality elevates the importance of simulation and its environment is critical to step up speed, capacity, availability, performance and reliability of design, engineering, manufacturing processes and systems.” The authors focus on their top 10 initiatives to address engineering and design COVID-19 needs, out of which “governance and supporting technologies” is identified to be the most challenging one in terms of time and cost to implement. Among others, the “increased volatility, economic pressures, changing suppliers, and more demanding customers increase the risk of missed purposefulness in design and simulation activities.” One of the keys for governing “increased volatility in design and simulation activities” is uncertainty quantification, and this is also a core topic of the article to be considered next.
In the contribution entitled “The need for speed to reconcile rationalism and empiricism in a data rich world”, Chris Smith states that “there are an increasing number of situations where vast quantities of data collected from asset-based sensors are not utilized effectively due to the lack of an interpretative framework”, and that “this drives the need for physical frameworks on which to base predictive capabilities.” The introductory part of the article particularly refers to a presentation given by Prof. Karen Willcox, who is with the Oden Institute for Computational Engineering and Sciences at the University of Texas in Austin, on the occasion of SC19: The International Conference for High Performance Computing, Networking, Storage, and Analysis at the Colorado Convention Center in Denver, CO, USA, in November 2019. In her presentation, Prof. Willcox points out that “big decisions need more than big data … they need big models too”. In particular, “big decisions must incorporate the predictive power, interpretability, and domain knowledge of physics-based models”. According to Prof. Willcox, the four important aspects in this context are that
- The dynamics of high-consequence applications are driven by complex multiscale multiphysics applications,
- There are high-dimensional parameters underlying the characterization of scientific and engineering systems,
- Data are sparse, intrusive and expensive to acquire, especially in the most critical regimes, and
- Uncertainty quantification (UQ) in model inference and certified predictions in regimes beyond training data play a critical role.
AdCo EngineeringGW bundles know-how from statistics, machine learning and engineering, that is, our “big models”, to provide innovative algorithms for physics-based UQ of systems with high-dimensional parameters in our software QUEENS. This is supplemented by our comprehensive expertise in the solution of complex multiscale multiphysics applications. Thereby, we enable our clients to harness the full potential that digital product development tools and processes can offer. For more information, please visit, for instance,
