MORESCO Corporation — Case Study

■Client Information

Company Name	MORESCO Corporation
Location	5-5-3 Minatojima-Minamimachi, Chuo-ku, Kobe
Business Description	Development, manufacturing, and sales of specialty lubricants, synthetic lubricants, materials, hot-melt adhesives, and energy device materials

We spoke with Sueyoshi and Fujii from the Research & Development Department.

Datachemical LAB is being utilized in the company's Research & Development Department. We spoke with Fujii (left) and Sueyoshi (right), who are leading the internal adoption of Datachemical LAB.

Q1. What themes are you using Datachemical LAB for?

MORESCO primarily develops products that perform functions such as lubrication, adhesion, and surface protection in the "interfacial domain" — the boundary where materials come into contact with one another. Within this scope, we are using Datachemical LAB for development projects related to cutting fluids, synthetic lubricants, hot-melt adhesives, and life science applications.

We particularly value the fact that Datachemical LAB is applicable not only to formulation-based themes but also to synthesis-based themes, with minimal constraints on data volume — enabling machine learning-based analysis on datasets ranging from small to moderate sample sizes. Even within our company, where product development workflows, methodologies, and daily experimental throughput vary across divisions, we feel we have been able to advance company-wide Materials Informatics (MI) adoption without limiting it to specific themes or business areas.

Q2. What were your reasons for selecting Datachemical LAB?

Before formally introducing MI, we organized internal study groups with a group of volunteers, using Professor Kaneko's textbooks as our learning materials. At the time, we were studying Python and machine learning in parallel, which placed a significant burden on employees, and the path to the most critical goal — actually driving MI adoption internally to improve product development efficiency — felt long and steep. Even though we understood that machine learning was a powerful tool, the resource investment required to educate developers to the point of practical use presented a very high barrier.

It was against this backdrop that we discovered Datachemical LAB, which makes Professor Kaneko's expertise accessible in a no-code environment. We immediately reported it to our supervisor at the time, and implementation followed without delay.

Q3. What benefits have you seen since introducing Datachemical LAB?

First and foremost, the fact that developers were freed from the need to learn Python meant we could move directly into exploring product development using Materials Informatics. We have been using Datachemical LAB since 2022, and looking back, we are confident that without it, we would never have reached where we are today — and it is possible that without being able to demonstrate any concrete results, the MI initiative itself might have stalled entirely.

It has also proven highly effective as an educational tool. During the early period of our MI journey, we were running study groups using Professor Kaneko's books. When those transitioned to study groups using Datachemical LAB — where participants could acquire knowledge while actually implementing machine learning hands-on — understanding of MI within the company advanced rapidly.

In terms of actual product development, cases have also begun to emerge: developers who started learning about MI at the same time Datachemical LAB was introduced have used Bayesian Optimization-based experimental design to improve the efficiency of new product development, and have been able to identify Pareto-optimal solutions that had previously gone undetected.

We feel that Datachemical LAB has given a significant boost to MORESCO's MI promotion efforts.

Q4. How do you feel about post-implementation updates and our support?

We have been using Datachemical LAB for nearly two years (as of September 2024), and the frequency of feature updates is truly remarkable. Through Datachemical LAB, we have been continually impressed by the pace of technological advancement in Materials Informatics and beyond.

Beyond the features themselves, the no-code machine learning capability alone was already enormously valuable — but we can also sense an ongoing commitment to evolving the user interface to become even more accessible to those new to Materials Informatics.

In terms of support, we receive responses to our questions within the same day, and there have been instances where our feature requests were incorporated into updates — giving us the impression that the team is genuinely committed to addressing user needs.

Q5. What are your plans for expanding the use of Datachemical LAB internally?

Our medium-to-long-term vision is to grow the number of personnel capable of conducting MI-driven product development to more than five times the current figure. However, our goal is not to increase the number of data scientists with a primary focus on data science — rather, we aim to develop "dual-skilled" developers who, first and foremost as MORESCO developers, are able to leverage Materials Informatics within a product development-centered role.

To that end, we plan to continue supporting developers in mastering Datachemical LAB, just as we have done so far, and to further accelerate the momentum of MI adoption across MORESCO.

← Back to Case Studies