From Weeks to a Day: Mitsubishi Chemical Group Accelerating Material Characterization
Mitsubishi Chemical Group (MCG) is a renowned leader in material technologies and solutions for various industries including the automotive industry. MCG performance polymers materials play a crucial role for taking this industry’s challenges such as comfort and well-being, electrification, connectivity, environmental constraints and resources preservation. To meet the ever-evolving demands of the industry and customers Mitsubishi Chemical Group is committed to support their partners with material characterizations used as input data for highly dynamic crash simulations using Finite Elements Analysis. The accuracy of these data is of particular importance for MCG’s high performance Thermoplastic Elastomers used in automotive safety applications (airbag).
Traditionally, MCG’s process of generating material data involved a combination of experimental tests and optimization runs. Engineers painstakingly sought to achieve a close match between the simulated behavior of the material and the observed results from experimental tests in a wide range of temperature and strain rates. This iterative approach required numerous optimization runs, resulting in significant time consumption for engineers.
In April 2020, MCG embarked on a collaborative journey with Neural Concept to explore alternatives to accelerate this time-consuming process and to improve the accuracy of the generated data. Through close cooperation, they together developed a powerful AI model able to accurately predict simulation results by leveraging a wide range of material laws as inputs.
Once trained, the AI model enables a reverse optimization process. Based on experimentally tested material data it automatically creates an optimized material law by matching the simulation results with the objective experimental curves. This new workflow delivers results within seconds, a dramatic improvement compared to the hours required for a single optimization run using traditional methods. While also improving the accuracy significantly.
What makes this collaboration even more impactful is that the Mitsubishi Chemical Group can now extend this workflow beyond standardized material laws to fully tailored solution according to unique customer demands. And to provide the required data within days.
We are now able to provide a qualified answer to our customers within a single day instead of several weeks with our previous traditional workflow.
Dr. Jan Klein, Head of Product Line EMEA & Global Product Line Manager Performance Polymers EMEA
Following a successful pilot project, Mitsubishi Chemical Group has fully implemented and deployed the Neural Concept solution in their experts' and designers' workflow. Leveraging the Neural Concept Cloud, they have achieved a new level of efficiency and accuracy, solidifying their position as pioneers in providing innovative materials and solutions.
About Mitsubishi Chemical Group (MCG)
Mitsubishi Chemical Group Corporation is a specialty materials provider with an unwavering commitment to lead with innovative solutions to achieve KAITEKI, the well-being of people and the planet. We bring deep expertise and material science leadership in core market segments such as mobility, health & life sciences, and electronics, enabling industry transformation, technology breakthroughs, and longer, more fruitful lives for us all. Together, around 70,000 employees worldwide provide advanced chemistry-based solutions to deliver the core elements of our slogan—Science. Value. Life.
About Neural Concept
Neural Concept develops Deep Learning algorithms for enhanced engineering. By speeding up R&D cycles, augmenting product performance, and solving their next engineering challenges, we revolutionize how our customers design their products. Our platform, Neural Concept Shape (NCS), brings intelligence to engineering workflows by merging data-science and deep learning into CAE/CAD workflows. The company was founded in 2018 in a top-notch AI research lab at EPFL in Switzerland. We are a team of 40 members fully committed to our vision, which is to revolutionize the future of industrial engineering with intelligence.
