Showa Denko K.K. (SDK) is first to introduce Machine Learning Operations (MLOps) for the efficient management of machine learning models deployed into artificial intelligence (AI) systems for materials design.
MLOps is a method and philosophy for integrating the development and operation of machine learning models. It involves continuous training of machine learning models and automating the machine learning process. This method also includes establishing tools and operational rules for collaborative development between data scientists and software engineers.
Machine Learning in Materials Development
These machine learning models can predict material properties based on formulations and the manufacturing-process conditions of materials.
This time, SDK has automated input of the latest data into computers that develop machine learning models. Also, it has automated data processing in these computers. The automation process has reduced the time required to build and operate machine learning models from five days to one day per month.
In addition, the introduction of MLOps enabled SDK to accelerate materials development by predicting material properties based on latest data.
AI Systems Promotes Efficiency
SDK utilizes AI systems for efficient materials development, such as exploring the optimal material formulation. Meanwhile, machine learning models deployed into the AI systems predict material properties from formulations or suggest formulations that improve material properties.
The machine learning process for managing AI systems includes inputting the latest data, data processing, and continuous training of machine learning models. Previously, data scientists had to input and process the latest data for themselves. However, these steps accounted for about 80 percent of the time required for the entire machine learning process.
These models deployed into AI systems are built specifically for each material. Therefore, before introducing MLOps, materials development required a lot of time and effort due to the necessary work specialized for each material.
SDK has installed programs to automate the input of the latest data and data processing into the company’s AI systems. This step aims to address issues caused by applying AI systems to the development of numerous materials in the company. It also targeted to operate machine learning models efficiently.
Moreover, the company has introduced technologies that would enable data scientists responsible for building machine learning models and software engineers responsible for building AI systems to develop systems collaboratively. This can be accomplished even if there are differences in the operating systems and programming languages they use.
Ahead of Competition
SDK has introduced MLOps ahead of its competitors to manage machine learning models efficiently. Therefore, it can reduce the time required to develop machine learning models and their operation. Thus, the company can improve prediction accuracy and stably operate dozens of AI systems. As a result, it can propose ideal materials to its customers promptly.
The Showa Denko Group will apply the fruits of basic research in AI and computational science to materials development. It will quickly provide solutions that solve its customers’ problems, thereby contributing to the development of a sustainable society.