Yokohama Rubber Generative AI System Accelerates Tire Development

The Yokohama Rubber Co., Ltd. has developed a proprietary generative AI system designed to search technical documents accumulated internally and generate responses from their contents. The company began full-scale operation of the system in August 2026. The solution is intended to give development personnel rapid and accurate access to technical information needed for decision-making across the tire development process, including material development. By making internal technical knowledge easier to retrieve and apply, the system is expected to support faster development activities while improving the quality and efficiency of decisions made during product development.

How the AI System Supports Tire Development

The Yokohama Rubber Co., Ltd. system uses Retrieval-Augmented Generation, or RAG, to connect employee questions with relevant information stored in internal technical documents. Development staff can enter questions based on specific objectives and circumstances encountered during development, after which the system searches the available technical information and generates responses grounded in the retrieved content. This approach is intended to reduce the time required to locate useful technical knowledge and provide development teams with information that is directly relevant to the issue or decision being considered.

AI Agent Improves Search Accuracy

An AI agent has also been incorporated into the system to better understand the intent behind questions submitted by development staff. Rather than relying on a single search operation, the agent can autonomously repeat a sequence involving search planning, information retrieval, and evaluation of the retrieved results. This iterative process is designed to improve the relevance and accuracy of the information used to formulate responses. For The Yokohama Rubber Co., Ltd., the capability provides a more structured way to use accumulated technical knowledge when addressing development-related questions.

Source Verification and Future Development

The system provides links to the original technical documents used as the basis for its responses, allowing development personnel to examine the underlying information and assess whether the generated answer is appropriate. This verification capability is important when technical information is used in development decisions because employees can review the source material rather than relying solely on an AI-generated response. The company plans to continue improving the system's capabilities while making greater use of internally accumulated data and knowledge. These efforts are intended to support the development of innovative products, processes, and services by The Yokohama Rubber Co., Ltd..

Full-scale operation began in August 2026, marking a broader application of generative AI within the company's technical development activities. By combining internal document retrieval, RAG, an autonomous AI agent, and direct access to source documents, the system brings several functions together within the development workflow. The approach can help employees find relevant technical information more efficiently while retaining the ability to verify the material supporting each response. As The Yokohama Rubber Co., Ltd. continues expanding the system, its accumulated data and knowledge are expected to play a greater role in future product, process, and service development.

Frequently Asked Questions

What is Yokohama Rubber's new generative AI system designed to do?
The system searches internally accumulated technical documents and generates responses based on relevant retrieved information. It uses Retrieval-Augmented Generation and an AI agent that can repeatedly plan searches, retrieve information, and evaluate results to improve response accuracy. Development personnel can use the system to access technical knowledge relevant to specific objectives and situations during tire development, including material development. The system also provides links to the original documents supporting its responses, allowing employees to verify the information before using it in technical decision-making and development activities.