Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • ADVERTISEMENT FEATURE Advertiser retains sole responsibility for the content of this article

Human–AI collaboration inspires tyre innovation

Simulations on the composition of tyre rubber provide data that HAICoLab uses to uncover new ways to boost tyre performance.© THE YOKOHAMA RUBBER CO., LTD.

At the heart of tyre technology lies the seemingly unattainable goal of providing the highest grip for safety while simultaneously presenting the lowest rolling resistance for good fuel economy and long wear. The level of performance offered by the humble modern tyre is a remarkable feat of research and engineering — not just in tyre design and structure, but also in the development of the complex composition and morphology of the rubber compound itself.

After more than a century of development, however, the next leap in tyre technology will require something more than ‘mere’ human ingenuity. Yokohama Rubber, a leader in tyre development since the 1920s, has now taken up this next-generation challenge by bringing artificial intelligence (AI) and big informatics to the design table.

“Yokohama has been developing technology to apply computational science and machine learning to rubber and tyre development for more than a decade,” says Masataka Koishi, head of the AI Laboratory at Yokohama Rubber. “The HAICoLab platform we announced in October 2020 is the culmination of this development. It’s a framework that emphasizes collaboration between humans and AI.”

Using AI to unshackle human inspiration

Short for ‘humans and AI collaborate for digital innovation’, HAICoLab is Yokohama’s solution to the other conundrum of this challenging R&D problem — how to exploit the human inspiration and ingenuity that allows us to look beyond the boundaries of what is known, while curbing the unconscious cognitive bias that routes our thinking and decision-making along well-trodden pathways.

Tyres have to maximize both grip for safety while minimizing lowest rolling resistance.© Marin Tomas/Moment/Getty Images

“We recognize that humans are indispensable for technological innovation, even when using AI,” says Koishi. “Human inspiration increases the chances of gaining insights that lead to radical innovations that are not just extensions of current knowledge.”

AI and machine learning can be used to rapidly and accurately predict tyre characteristics based on past measurement data and computational science simulation data, which can greatly speed up the development process and drive incremental innovation. Since AI depends on learning data, however, it cannot be readily applied to search unexplored areas that lack learning data, and so it is not suited to radical innovation.

HAICoLab is used to create and collect large amounts of data from measurements and simulations. Then, based on hypothetical conditions set by humans, it predicts, analyses and searches for improvements. This process generates new knowledge, which human engineers use to set new hypotheses. By drawing on principles from behavioural psychology and behavioural economics, the formulation of hypotheses is framed to eliminate biases that could hinder new discoveries.

“The knowledge and inspiration gained from AI and the elimination of cognitive bias give us the direction and ideas, our hypotheses,” says Koishi.

It starts with the rubber

Tyre rubber is not just rubber. It is a painstakingly engineered compound consisting of a rubber polymer and a filler such as carbon black or silica, combined in a complex morphology defined by the properties, dispersion, size, quality and modification of each of the components at the macro and micro scales.

“The number of factors to be considered is enormous, and so computational simulations such as molecular dynamics simulations and even quantum calculations are needed,” Koishi says. “Machine learning can speed up the examination of material properties and a bird’s-eye view of the vast array of multidimensional data that can be used to determine the direction of development.”

HAICoLab has found new insights that will improve tyre performance through altering the make-up of tyre rubber.© THE YOKOHAMA RUBBER CO., LTD.

HAICoLab has already yielded new insights that will improve tyre performance, revealing that smaller filler particles combined with a thin ‘bound rubber’ layer that forms on the filler particle surface due to interaction with the rubber polymer can lower rolling resistance and boost wear performance. Molecular dynamics simulations indicated that the small filler particles increase rigidity while the thinner bound rubber layer reduces energy loss.

“The properties required of tyres are diverse, including rolling resistance, steering stability, ride comfort, durability, wear resistance and braking performance,” says Koishi. “There is a trade-off between all of these characteristics, making it difficult to improve them all at the same time. We’re applying HAICoLab to address this challenging problem.”

Using AI to accelerate, not replace, human R&D

This new development platform is a model for collaboration between humans and AI that, far from sidelining the vast expertise of its human collaborators, unlocks the full potential of both. Using AI to search through the vast amounts of data from many simulations across an enormous multidimensional design space will make it possible to derive the morphological design factors critical to achieving the desired performance in a much shorter time with a higher degree of objectivity and quantitative accuracy.

Not only does this greatly improve the precision of material exploration and reduce the hours spent in physical prototyping, it also gives the human engineers more time and space to innovate and develop themselves. “The ‘HAICoLab loop’ greatly improves the probability of radical innovation, as well as promoting the growth of our engineers,” Koishi says.

Find more about the products:

Related Articles


Quick links