Yokohama is creating an AI agent for tire development
Yokohama is collaborating with dotData, a U.S.-based artificial intelligence and advanced data analytics startup, to develop an AI agent that will assist engineers in formulating hypotheses, interpreting data, and making decisions during the development of new tires. The joint project aims to evaluate the agent’s capabilities prior to its full-scale implementation in research and development, the tire company reported.
The new program will extract relevant data from technical documentation, including test reports, and deliver information and analytics aligned with the objectives of each research project. Yokohama emphasizes that the AI is not intended to provide a single correct answer. Instead, it will employ a Human-in-the-Loop model, in which humans actively participate in the algorithm’s workflow by reviewing and refining its outputs. This way, employees retain responsibility for decisions made, including regarding potential risks, and the new approach is expected to support their professional growth through continuous collaboration with AI.
Yokohama adds that during research and development, a large volume of data is generated through testing, simulation, and other methods. To formulate new technical hypotheses, engineers must interpret information and make decisions based on specialized knowledge and context—including project goals, existing constraints, and prior experience. Abductive reasoning plays an important role in this process, a method in which several hypotheses are formed based on observed phenomena, and the one considered the most likely explanation is selected. Humans have an advantage in hypothesis generation due to their experience and knowledge; however, 'past successes and established assumptions may narrow the range of possibilities considered,' and this is where artificial intelligence should help.
Yokohama has been developing the AI-based HAICoLab platform since 2020 to accelerate innovation in the development of products, manufacturing processes, and services, and the collaboration with dotData aims to expand the application of artificial intelligence in R&D.
"Within the HAICoLab project, we leverage the complementary strengths of humans and AI," noted Motodiji Fujita, head of Yokohama's research and development department, adding that "AI agents will give engineers access to perspectives beyond their own experience and enable them to evaluate hypotheses from multiple angles."
"Through this initiative, we aim to enhance our R&D capabilities while simultaneously supporting the further professional growth of our engineers," he stated.