Yokohama Rubber AI Agent Development Begins in R&D

Yokohama Rubber AI Agent Development begins as Yokohama Rubber Co., Ltd. and U.S.-based dotData jointly work on an AI agent for manufacturing research and development. Announced on September 16, the initiative is intended to support engineers as they form hypotheses, interpret findings, and make decisions during R&D activities. Before the technology is introduced into actual R&D work, the companies will verify its required functions and technical feasibility. The project builds on Yokohama Rubber’s existing use of artificial intelligence in innovation activities and focuses on extending those capabilities through an agent-based approach designed to assist engineers while keeping final interpretation and decision-making with human specialists.

AI Agent Development for Manufacturing R&D

The jointly developed system will combine several sources of technical knowledge and data used in manufacturing research. These include technical documents such as design information, experiment reports, and research reports, along with knowledge accumulated by engineers. The system will also use features derived from measurement and experimental data through dotData’s automated feature engineering technology. By bringing these inputs together, the agent is intended to provide engineers with a broader basis for considering research questions. The companies are therefore focusing not only on generating AI outputs, but also on integrating technical information and data-derived insights into an engineering-oriented R&D workflow.

Hypotheses, Technical Perspectives, and Risks

For each research theme, the AI agent will consider factors including the objectives, subjects, and constraints involved in the work. Based on those conditions, it will present multiple hypotheses for engineers to examine, together with comments from different technical perspectives and potential risks that should be considered during verification. This approach is designed to support the exploratory stages of manufacturing R&D, where engineers must evaluate different possibilities and determine which findings warrant further investigation. Rather than replacing engineering judgment, the system is structured to provide additional perspectives that can help organize and broaden the evaluation of research results.

Human-in-the-Loop Engineering Approach

The system will use a human-in-the-loop approach, meaning engineers remain responsible for the final interpretation of findings and the decisions that follow. This operating model places the AI agent in a supporting role within the R&D process rather than giving it independent authority over engineering conclusions. Yokohama Rubber Co Ltd is expected to use the joint development process to determine which functions are technically feasible and appropriate before applying the technology to actual manufacturing research themes. The companies will first conduct verification work, creating a basis for assessing how the agent can contribute to practical engineering activities.

Yokohama Rubber’s HAICoLab Initiative

Since 2020, Yokohama Rubber Co Ltd has promoted HAICoLab, an AI utilization framework intended to drive innovation across products, processes, and services while developing human resources through collaboration between people and AI. The company has previously used dotData as a tool supporting these efforts. The new joint development represents an effort to advance that existing AI utilization framework toward an AI-agent-based approach. Its focus is specifically on applying the technology to actual manufacturing R&D themes and strengthening how engineers can use AI within research activities.

Industry Impact & Outlook

The development could affect manufacturing R&D by moving AI assistance beyond isolated analytical tasks toward a workflow that connects technical documents, accumulated engineering knowledge, and experimental data. For Yokohama Rubber, the immediate significance is the opportunity to test whether an AI agent can provide useful hypotheses and technical perspectives while preserving human control over engineering decisions. The verification phase will determine the functions and technical feasibility required for practical deployment, making the next steps dependent on those results. More broadly, the project illustrates how United States-based AI technology can be incorporated into manufacturing research through collaboration with established engineering organizations.

Frequently Asked Questions

What is Yokohama Rubber developing with dotData?
The companies are jointly developing an AI agent designed to support engineers during manufacturing research and development activities. The system will combine technical documents, accumulated engineering knowledge, and features derived from measurement and experimental data. It is intended to present multiple hypotheses, technical perspectives, and potential verification risks based on each research theme’s objectives, subjects, and constraints. Engineers will remain responsible for interpreting the findings and making final decisions. Before deployment in actual R&D activities, the companies will verify the agent’s required functions and technical feasibility.