a.s.ist Selected for Resonac's Business Co-creation Program "OnStage"

a.s.ist Inc. (hereinafter referred to as a.s.ist) is pleased to announce its selection for "OnStage," a business co-creation program hosted by Resonac Corporation.

"OnStage" is a business co-creation program that aims to create new businesses that solve social issues through co-creation between Resonac and external companies. In this program, Resonac employees and external companies form co-creation teams and proceed with verification activities for commercialization, receiving mentoring and support for utilizing Resonac's assets.

a.s.ist has been working on automating data analysis in materials development and physical property experiments using statistical machine learning technology. Through this program, we will leverage our automated data analysis technology and experimental condition optimization technology to co-create towards the advancement of the "synthesis, measurement, analysis, judgment, and next condition determination" cycle in materials R&D.

OnStage Selection Details for Our Company

Theme: Realization of Lab Automation for Materials R&D Centered on Automated Data Analysis Technology

Selection Details: We will conduct verification for the advancement of data utilization and analysis operations in Resonac's research and development fields, and the efficiency of the R&D cycle, by utilizing a.s.ist's automated analysis technology for spectral data and material analysis data using Bayesian inference, etc., as well as our experimental condition exploration and optimization technology.

About OnStage

"OnStage" is a business co-creation program hosted by Resonac's R&D base, "Co-creation Stage." Resonac and external companies such as startups combine their respective technologies, knowledge, and assets to aim for the creation of new businesses that solve social issues.

In this program, co-creation teams composed of Resonac employees and selected companies will be formed, and will proceed step-by-step with hypothesis verification of customer issues, construction of value propositions and solutions, prototyping, PoC, and business plan development.

OnStage Special Website: https://unidge.co.jp/project/onstage

Background of Selection

In materials development, there are situations that require advanced expertise and significant time in the series of processes including designing experimental conditions, analyzing measurement data, interpreting results, and determining the next experimental conditions. In particular, spectral and measurement data analysis such as XRD, XPS, FT-IR, and NMR*1 often depends on the experience of the analyst for judgments such as the number of peaks, peak shape, background, and model selection*2, which can lead to issues with reproducibility and efficiency.

Based on research knowledge cultivated at the University of Tokyo, a.s.ist has developed white-box AI*4 technology utilizing statistical methods such as Bayesian inference and sparse modeling*3. This supports automated analysis of materials and physical property experimental data, feature extraction*5, and optimization of experimental conditions, promoting DX in research and manufacturing sites.

Through this OnStage selection, a.s.ist aims to create new value by working with Resonac to advance data analysis and decision-making in materials R&D.

Future Developments

During the OnStage program period, a.s.ist will verify the applicability of our technology while considering the challenges and needs in Resonac's R&D sites.

Specifically, we plan to conduct technical verification for automated processing of material analysis data, proposal of next conditions based on experimental results, and overall efficiency improvement of the R&D process. In the future, we aim to realize solutions that contribute to reducing personalization in materials development sites, improving the efficiency of analysis operations, and accelerating the R&D cycle.

a.s.ist will continue to promote the advancement and social implementation of data utilization in the materials development and manufacturing industries by leveraging statistical machine learning and white-box AI.

About a.s.ist Inc.

a.s.ist is a startup company that works on advancing materials development and physical property experiments by utilizing advanced data analysis technologies such as Bayesian inference, sparse modeling, and white-box AI, based on research knowledge cultivated at the University of Tokyo.

In particular, we specialize in the automated analysis of spectral data and material analysis data such as XRD, XPS, FT-IR, and NMR, and support the efficiency and improved reproducibility of highly specialized analysis tasks such as peak separation, parameter estimation, and feature extraction.

In addition to developing and selling automated analysis software, we promote the advancement of the entire "synthesis, measurement, analysis, judgment, and next condition determination" cycle in materials R&D through experimental condition optimization, integrated utilization of R&D data, automatic report generation, and provision of various DX solutions.

Currently, a.s.ist is developing and implementing automation technologies aimed at realizing lab automation and autonomous materials exploration*6 in R&D sites, with automated spectral analysis and experimental condition optimization technologies as its core.

Company Name: a.s.ist Inc. Representative Directors: Yuui Hayashi / Ryota Moriguchi Location: 5-17-24 Iko, Adachi-ku, Tokyo R&D Base: Room 606, Tokei Techno Plaza, 5-4-6 Kashiwanoha, Kashiwa-shi, Chiba Established: June 2021 Business Activities: Analysis software development, anomaly detection, automatic report generation, algorithm development URL: https://www.a-s-ist.com/

List of Notes

*1 XRD (X-ray Diffraction), XPS (X-ray Photoelectron Spectroscopy), FT-IR (Fourier Transform Infrared Spectroscopy), NMR (Nuclear Magnetic Resonance)

Representative analysis methods for investigating the components, molecular structure, crystal structure, and surface state of materials and chemical substances.

*2 Model Selection

Choosing the mathematical model or calculation method that is most suitable for the analysis target.

*3 Sparse Modeling

A technique for analyzing important information by selecting only it from a large amount of data, which can yield easily interpretable models.

*4 White-box AI

Technology for building AI in a way that humans can easily understand and explain the reasons for its conclusions.

*5 Feature Extraction

Technique for extracting only important features from a large amount of data and converting them into a form that is easy to analyze.

*6 Autonomous Materials Exploration

Technology that efficiently explores new materials by utilizing AI and robots to automatically repeat the cycle of experimentation, analysis, and determination of the next experimental conditions.

Inquiries Regarding This Release

a.s.ist Inc. Public Relations Email: [email protected]