YOKOHAMA's Technology
Yokohama Rubber aiming for digital innovations with HAICoLab, a proprietary framework for utilizing AI that facilitates collaborative efforts by humans and AI
Basic Technology
In recent years, AI (artificial intelligence), known for its ability to process vast amounts of data, has become indispensable in various industries. However, here at Yokohama Rubber, we realize that inspiration and creativity from humans is indispensable for innovation in previously unexplored areas that lack the data needed by AI. Accordingly, in 2020, we introduced HAICoLab, a proprietary AI utilization framework that we are using to develop innovative products, services, and processes through collaboration between humans and AI. We are now using HAICoLab to develop various technologies.
HAICoLab synopsis
The HAICoLab process begins with the collection of data based on hypothetical conditions set by humans and then applying AI to predict, analyze, search for and generate useful information (= features). Using the analogy of aiming for a destination from your current location, the features referred to here correspond to the direction (= factors) and distance (= quantitative value) that must be traveled to reach the desired destination. Humans then make decisions based on their interpretation and consideration of the features obtained by AI . This concerted effort of AI and human generates greater awareness and inspiration that produces actions arriving at the target destination. In addition to Yokohama Rubber’s AI technologies, HAICoLab makes use of behavioral psychology and behavioral economics to reduce preconceived notions and human biases that can affect the use of AI.
Yokohama Rubber is applying its HAICoLab concept to develop proprietary AI systems that are being used in its development of various new tires. Examples of these AI systems include one that instantly predicts the physical property values of rubber and the values of key tire characteristics, and another being used to create rubber compounds that meet specified physical property values. In May 2024, Yokohama Rubber’s “R&D on HAICoLab, data utilization through concerted effortbetween humans and AI” received the 36th Society of Rubber Science and Technology, Japan Award.
Example of HAICoLab’s use in rubber mixing process
AI’s use in the rubber mixing process is a good example of the HAICoLab concept of collaborations between humans and AI. The tire manufacturing process begins with the mixing of rubber, the main raw material in tires, and the quality of the mixed rubber is a key factor affecting tire performance. Rubber quality is affected by various factors in the mixing process, including temperature and humidity, mixer rotation speed and power, and mixing time length. To determine which elements in these factors contribute the most to quality stabilization, Yokohama Rubber utilized dotData, an AI tool that automatically discovers and extracts features. By inputting data from past mixing processes into dotData and then applying uniquely multifaceted human interpretations of the obtained features received from materials engineers and data scientists, Yokohama Rubber discovered the features that are most important for realizing quality stabilization. The key to success was the facilitation that led to multifaceted interpretations. Mass production of prototypes with different mixing specifications based on the results of the above process have proven its efficacy.
Yokohama Rubber will expand the use of HAICoLab from technical areas as it aims for innovations to be realized by collaborations between humans and AI that merge human inspiration and creativity with the enormous data processing capability of AI.
