Yokohama Rubber, dotData Partner on AI Agent for Manufacturing R&D
Yokohama Rubber and dotData announced a joint development initiative to build an AI agent for manufacturing research and development. The system will support Yokohama engineers in forming hypotheses, interpreting findings and making decisions, the company said.
The companies will validate the system’s capabilities and technical feasibility before any full-scale deployment.
The system will combine technical documents, engineers’ accumulated expertise and features derived from measurement data. Technical documents may include design specifications and experimental reports. The system will use dotData’s Signal Intelligence technology to derive features from measurement data.
The goal is to provide information and insights relevant to each R&D project’s objectives, circumstances and constraints.
The AI agent uses a human-in-the-loop framework that keeps engineers responsible for decisions. Rather than provide a single answer, the system will present multiple possibilities, perspectives and potential risks to broaden engineers’ thinking.
Yokohama AI Agent Supports R&D
Manufacturing R&D generates diverse data from experiments, simulations, production and evaluation. Developing new technical hypotheses requires engineers to interpret findings within the context of objectives, experimental conditions, technical constraints and prior experience.
R&D rarely produces a single predetermined answer. As a result, abductive reasoning plays an important role in forming and refining multiple hypotheses based on observed phenomena.
Experience can support this work, but established assumptions can also narrow possibilities. Yokohama Rubber and dotData say AI can broaden perspectives by using data and organizational knowledge to augment human judgment.
The companies intend to develop a system that understands a research project’s objectives, targets and constraints. It will find relevant information and support engineers across several areas.
System Combines Documents and Engineering Knowledge
The AI agent will reference design specifications, experimental reports and research summaries. It will also incorporate accumulated engineering knowledge. This information will support evaluations grounded in the background and history of individual R&D projects.
Using dotData’s Signal Intelligence technology, the system will also explore patterns in measurement and experimental data. The analysis will not be limited to previously identified factors.
Data-derived features will provide quantitative insights. When combined with design documents and expert knowledge, they may reveal relationships that are difficult to identify through experience alone.
The companies expect these insights to support development of new hypotheses.
AI Agent Presents Multiple Perspectives
The system will provide input from materials, design, analysis, production and evaluation perspectives. It will also introduce viewpoints that challenge existing hypotheses.
The AI agent will raise questions and identify potential risks to help engineers test assumptions, including: “Could this data support another interpretation?” and “Under what conditions would this relationship no longer hold?”
Engineers will remain responsible for final interpretation and decision-making.
They will develop hypotheses based on the information presented and test them through experiments or analysis. Engineers can then incorporate those results into further discussions with the AI agent.
Through dialogue, engineers can communicate analytical objectives and areas of focus. That input will guide Signal Intelligence processes and settings.
The approach is intended to allow non–data science specialists to explore relevant features as needed.
Development Builds on HAICoLab
Yokohama Rubber launched its HAICoLab AI utilization framework in 2020. The framework promotes innovation in products, processes and services through human-AI collaboration.
The company has used dotData as a foundational technology. It has applied multiple interpretations to automatically generated features in tire design and rubber mixing processes to develop new insights and hypotheses.
The joint development will expand those efforts through an AI agent.
Engineers will interpret and evaluate the agent’s suggested features or observations using their specialized knowledge and experience. Repeating this process can help engineers incorporate new perspectives and broaden the mental models they use for analysis and decision-making.
The companies aim to transform R&D processes while strengthening engineers’ creativity and judgment.
HAICoLab creates a continuous cycle of innovation in products, processes and services while supporting human development. The process begins by enhancing AI with data and knowledge accumulated within the company.
Engineers then use metacognition to develop hypotheses. They apply their expertise to interpret and evaluate the AI’s findings.
Companies Outline Future Development
Yokohama Rubber said it will continue applying the HAICoLab framework through AI agents in manufacturing R&D projects. The company aims to advance processes by combining human creativity with data-driven insights.
dotData will continue developing technology that combines quantitative context from automated Signal Intelligence with AI agents across manufacturing R&D.
The company plans to support diverse research topics, data formats and stages of development. Its goal is to expand human-AI collaboration across specialized business domains.