DUNLOP (Company name: Sumitomo Rubber Industries, Ltd. [1] and Fujitsu Limited [2] today revealed that they have actually collectively established an innovation, an AI surrogate design, to anticipate tire efficiency with high precision and in a brief time utilizing AI, verifying its efficiency in a current evidence of idea. This innovation, established as part of DUNLOP’s long-lasting management technique for style digital change, was used to the structural analysis of tire contortion when in contact with the roadway surface area. As an outcome, the analysis time was substantially minimized by roughly 90% from about 45 minutes to about 5 minutes, while attaining analysis of around 600,000 components (meshes).
Based upon the outcomes of this evidence of principle, both business will continue with the advancement of a style assistance tool for tire style, going for its useful application at DUNLOP by April 2027. This will allow DUNLOP to speed up data-driven advancement and promptly supply premium tires with improved security and ecological efficiency to the marketplace.
This innovation is created to operate on “FUJITSU-MONAKA,” [3] a next-generation Arm-based CPU established by Fujitsu that pursues both high efficiency and energy effectiveness. Progressing, both business intend to enhance reasoning speed, precision, and power effectiveness by beginning confirmation of this innovation on a model of FUJITSU-MONAKA by December 2026.
Background
In production, CAE (Computer Aided Engineering) analysis [4]which mimics the habits of items and structures to assess efficiency and security, needs a significant quantity of computational time due to the increasing efficiency and intricacy of items.
In tire style, FEM (Finite Element Method) analysis [5]a kind of CAE analysis, is typically utilized. While increasing the variety of components by fine-tuning the mesh enhances precision, it likewise increases calculation time and associated advancement expenses, demanding a balance in between precision and computational load. Analysis needs specialized understanding, and protecting experienced engineers is likewise a difficulty.
To resolve these problems, both business established an AI surrogate design that quickly forecasts services to the governing formulas of FEM utilizing collected FEM analysis results as training information.
Evidence of Concept Results
Leveraging DUNLOP’s tire style competence and real style information, together with Fujitsu’s AI innovation, both business collectively established an AI surrogate design based upon the Graph Neural Network (GNN) [6] algorithm. They performed an evidence of principle for tire structural analysis, concentrating on examining contortion habits and contact qualities such as contact shape and pressure circulation when a tire touches with the roadway surface area.
As an outcome, calculation time for the analysis was minimized from roughly 45 minutes to about 5 minutes. The innovation anticipated the contact shape in between the tire and the roadway surface area with a high typical precision of 87.7% compared to FEM analysis. This innovation will make it possible for much faster decision-making and enhance expenses, in addition to enhancing efficiency, by enabling the decision of tire structure and product specs in less procedures and a much shorter time, which formerly needed several style procedures.
A part of these outcomes will exist at the 31st Computational Engineering Conference, held from June 3, 2026.
Figure1: Proof of Concept Image
Figure 2: Relationship in between precision and calculation time in FEM analysis
Future Plans
Both business will start confirmation of this AI surrogate design utilizing a model of FUJITSU-MONAKA by December 2026, intending to enhance reasoning speed and power effectiveness. They will likewise broaden the application series of tire structural analysis and establish a style and advancement assistance tool that can be straight utilized by designers without specialized understanding. DUNLOP intends to start useful operation by April 2027. DUNLOP intends to understand its vision of “Continuing to Provide ‘New Experiential Value’ Born from Rubber to Everyone” under the Long-term Corporate Strategy “R.I.S.E. 2035.” [7] Through this co-creation with Fujitsu, DUNLOP will even more improve its exclusive Rubber and Analytical Technology, enhancing its technological abilities and speeding up development. By doing so, DUNLOP will implement its Purpose: “Through development we will develop a future of happiness and wellness for all.” and continue to provide brand-new worth to society.
Fujitsu will utilize this effort to promote horizontal implementation to massive FEM analysis in the vehicle and other production markets. Moving on, Fujitsu will add to enhancing advancement and promoting carbon neutrality through enhanced power performance in the production market by establishing an AI reasoning platform integrating FUJITSU-MONAKA and GNN, and using it on the AI platform “Fujitsu Kozuchi.” [8]
Hallmark Information
All business or item names pointed out herein are hallmarks or signed up hallmarks of their particular owners.
[1] Sumitomo Rubber Industries, Ltd.:Head Office: Kobe City, Hyogo Prefecture, Japan; President and CEO: Yasuaki Kuniyasu.
[2] Fujitsu Limited:
Head Office: Kawasaki City, Kanagawa Prefecture, Japan; Representative Director and President: Takahito Tokita.
[3] FUJITSU-MONAKA:
A processor based upon the Arm direction set architecture, making use of advanced 2-nanometer innovation. It accomplishes high power effectiveness through Fujitsu’s exclusive innovations such as its self-designed microarchitecture and ultra-low voltage innovation. It promotes the facility of an environment where efficiency can be made the most of through industry-standard software application compatibility through partnership with the OSS neighborhood. These brand-new innovations used to FUJITSU-MONAKA are based upon outcomes gotten from a task funded by the New Energy and Industrial Technology Development Organization (NEDO).
[4] CAE (Computer Aided Engineering) analysis:
An engineering approach that adds to reducing advancement time, lowering expenses, and enhancing quality by mimicing and examining the style and efficiency of items on a computer system. FEM (Finite Element Method) is among the representative mathematical analysis approaches that make up CAE.
[5] FEM (Finite Element Method) analysis:
A mathematical analysis approach extensively utilized for examining strength and contortion. It includes dividing a constant item, such as a structure, into various little areas and numerically resolving them by using physical laws to each.
[6] Chart neural network (GNN):
An AI design efficient in straight managing graph-structured information (such as meshes) and discovering the interactions in between nodes.
[7] Long-lasting Corporate Strategy “R.I.S.E. 2035.”
[8] Fujitsu Kozuchi:
A group of AI services (platform) that incorporates research study and advancement with service, based upon Fujitsu’s innovative AI innovations. Its function is to rapidly advertise high-demand innovations from the research study phase and continually enhance and boost them.
About DUNLOP
DUNLOP (Company name: Sumitomo Rubber Industries, Ltd.) intends to understand its vision of “Continuing to Provide ‘New Experiential Value’ Born from Rubber to Everyone” under the Long-term Corporate Strategy “R.I.S.E. 2035.” Leveraging its core strength in Rubber and Analytical Technology, the Company establishes and offers product or services throughout a vast array of fields, consisting of tires, sports, and commercial items. With DUNLOP as its worldwide core brand name, the Company develops brand-new worth in varied locations such as movement, sports, commercial facilities, and health care. Assisted by its Purpose: “Through Innovation We Will Create a Future of Joy and Well-being for All,” DUNLOP adds to the awareness of a sustainable society. Learn more: https://www.srigroup.co.jp/english/
About Fujitsu
Fujitsu’s function is to make the world more sustainable by constructing rely on society through development. As the digital improvement partner of option for clients around the world, our 100,000 staff members work to deal with a few of the best obstacles dealing with mankind. Our series of services and options make use of 5 crucial innovations: AI, Computing, Networks, Data & & Security, and Converging Technologies, which we unite to provide sustainability change. Fujitsu Limited (TSE:6702) reported combined earnings of 3.5 trillion yen (US$ 23 billion) for the ended March 31, 2026 and stays the leading digital services business in Japan by market share. Learn more: global.fujitsu
Press Contacts
