Solving Material Development Bottlenecks
With AI to Enhance Industrial Capabilities
and Build a Sustainable Future

Climate change, resource exhaustion and environmental pollution are all challenges that we face today. For example, plastic does not break down in the natural environment and, if things are left as they are, it is predicted that the amount of plastic waste in the oceans by 2050 will exceed the amount of fish.^*1

One way to solve these challenges is to develop sustainable alternatives. Materials Informatics (MI) from our Materials Solutions Division offers solutions for the fast-paced development of materials. We are able to achieve higher efficiencies by using AI in processes that have traditionally relied heavily on experienced researchers, including research, design and simulation.

This significantly reduces the effort and cost, and most importantly the time, required to develop new materials. And it also plays a part in delivering new materials to our core operations in a speedy manner.

With the adoption of SDGs by the United Nations in 2015, the development of new materials is anticipated to help resolve environmental issues. This development is not only playing an important role in the realization of a sustainable and better world, but is also becoming essential for the development of advanced industries. The need for Materials Informatics continues to grow due to these trends.

The Japanese government is also paying close attention to these issues, and is expected to support the resolution of various social issues through, among other things, SDG No. 9 of "Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation."

In recent years, the Japanese government has also been conducting studies aimed at improving the country's capability to innovate in materials, in response to the growing importance of "materials" (including substances, raw materials and devices) to industry and innovation.

*1

From the World Economic Forum Report (2016)

AI Helps Reduce Waste and Speed Up the Development of Materials

Let's compare the use of AI in the development of materials to the traditional process (Fig.) The initial step of setting the target performance of a material is the same as before. It's the next step that is the big difference. Traditionally, humans handled a series of processes that involved studying similar development cases, creating designs and running simulations on candidate materials.

It relied on the experience and judgment of accomplished experts.

On the other hand, Materials Informatics uses AI, which relies on data-based analysis rather than the expertise and judgment of experienced personnel. This enables the discovery of the unknown, beyond even the imagination of experienced personnel.

Furthermore, in the past a lot of experiments were carried out blindly on the premise that "you won't know until you try it." However, by creating plans for experiments using AI, it is now possible to conduct only those experiments that have a high degree of confidence. As a result, the number of experiments can be significantly reduced.

The analysis uses previous experimental data from customers, but this can also be complemented with data extracted from patents, research papers and images.

CO₂ Is Also Reduced in Proportion to the Savings in Development Time

Yasuyuki Tamai of the Materials Solutions Division explains the effects of introducing Materials Informatics.

"Although it is difficult to generalize because so much depends on the conditions, there are some cases where the use of MI has reduced the number of experiments to one quarter of those previously carried out. This means that only a quarter of the amount of material and a quarter of the human resources and costs are needed; only a quarter of the waste is generated; and on top of all that, only a quarter of CO₂ emissions are produced."

Reducing CO₂ in manufacturing facilities is a challenge for all industries. Using Materials Informatics to reduce the number of experiments will not only allow us to reduce emissions from the energy (heat and electricity) used in those experiments (Scope 1 & Scope 2^*2), but will also enable reductions in emissions across the entire supply chain, from the procurement/transportation of raw materials to the disposal of waste (Scope 3^*2).

MI has received praise from customers, and the following type of feedback is often provided.

• "We have been able to streamline and accelerate our R&D process."
• "It offered us compatibilities and combination options that we would never have thought of with human resources alone"
• "We were able to use AI to verify that the knowledge we had accumulated was correct."

These results show the usefulness of Materials Informatics, where AI has replaced areas that used to rely on the experience and judgment of accomplished chemists. Yuya Nogawa, leader of the MI Sales Group, says:

"The service is offered as a subscription and is delivered via the cloud. This minimizes the work and cost of deploying large-scale systems, which lowers the obstacles to deployment for a wide range of customers, including small and medium-sized businesses."

*2

From "Green Value Chain Platform" on the Ministry of the Environment (MOE) website

Scope 1:

A business's own direct emissions of greenhouse gases (from the burning of fuel and industrial processes)

Scope 2:

Indirect emissions associated with the use of electricity, heat and steam supplied by other companies

Scope 3:

Indirect emissions other than those listed above (emissions from other companies related to the activities of the business)

Meeting Customer Expectations With High Analytical Capacity and the Ability To Solve Problems

Data scientists provide support for analysis conducted using Materials Informatics. In addition to Hitachi High-Tech, the Public Systems Division at Hitachi, Ltd. is also involved in providing expert support. Our data scientists have a variety of backgrounds, including AI and chemistry experts. They are able to handle a wide range of analytical requests, complementing each other with their respective areas of expertise.

The role of the data scientist is explained by Kouta Omobayashi, an engineer.

"Of course you need to have good analytical skills, but what's more important than anything is your ability to solve problems. First, you need to listen to each of the customer's problems and concerns. I am always conscious that we need to discuss not just the issues themselves, but also to make an effort to understand the background as to why such issues have occurred."

Risa Kawanaka, a marketing manager who observes the activities of data scientists, adds:

"Customers have told us how other companies bluntly refused their requests, saying they were "impossible," but Hitachi High-Tech were kind enough to listen and suggest solutions. We have received good ratings from customers for the way we listen carefully to all projects, identify the issues and come up with solutions."

Building a Bridge to Green Transformation

Materials Informatics was originally started in 2011 in the United States under the Obama administration. Japan has also been promoting cooperation between the public and private sectors as an effective means of developing sustainable industries and resolving social issues with the help of the manufacturing industry.

Hitachi, Ltd., part of the Hitachi Group, started to offer MI services in 2017. Hitachi High-Tech also became involved and the full-scale commercialization of the services began in FY2021. Yuya Nogawa, MI Sales Group Leader, explains Hitachi's strengths:

"The Hitachi Group has been providing many solutions to customers in various industries across the globe. Hitachi High-Tech also develops and manufactures analytical and measurement equipment in addition to Materials Informatics systems. Having a system in place that can offer a total service package, from various types of equipment through to analytical solutions, is a strength that other IT vendors don't possess."

We will continue with this.

"Materials Informatics is now between the introduction and growth stages. First, we will support digital transformation (DX) at manufacturing sites, and we hope to develop this into green transformation (GX) with a view to solving global environmental issues."

Hitachi High-Tech Group has formulated five materialities (key topics) for solving social issues, based on SDGs.

1. Contributing to a Sustainable Global Environment
2. Contributing to Healthy, Safe, Secure Lives
3. Contributing to the Sustained Development of Science and Industry
4. Securing a Sound Foundation for Running the Company
5. Developing and Utilizing Diverse Human Resources

Materials Informatics can use 3) as a starting point to contribute to 1) & 2), and in addition, it can help with 4) and 5) by cutting down development time through the use of AI. Materials Informatics is expanding its field of activity as it solves social issues from the dual perspective of both promoting industry and also ensuring its coexistence with the global environment.

• Hitachi High-Tech Group Materialities