AIZOTH Inc. participated in the 57th Autumn Meeting of the Society of Chemical Engineers, Japan (SCEJ), held at Hiroshima University’s Higashi-Hiroshima Campus from September 14 to 16, 2026. Through a research presentation and a luncheon seminar, we shared our work on applying AI and machine learning to chemical engineering research and explored opportunities to accelerate R&D with researchers and engineers from academia and industry.
Research Presentation: Advancing Chemical Engineering with AI
On September 14, we presented “Advancing Chemical Engineering Research through an Integrated AI Platform” in the session “Frontiers of Data-driven Research and Development” (ST-21).
Our presentation focused on how AI, cheminformatics, and optimization techniques can improve the efficiency of experiments, data analysis, and the exploration of experimental conditions. The session also featured research on laboratory automation, machine learning, and data-driven approaches to materials and process development, providing an opportunity to exchange insights with researchers working at the forefront of AI applications in chemical engineering.
Luncheon Seminar: Accelerating R&D with Multi-Sigma
On September 15, we hosted a luncheon seminar titled “Ultra-Efficient Chemical Engineering Research with the AI Analytics Platform Multi-Sigma: Achieving Innovative Discoveries with the Minimum Number of Experiments.”
The seminar attracted more than 80 attendees from universities, public research institutions, and industry, reflecting strong interest in practical applications of AI across chemical engineering and related fields.
We demonstrated how our no-code AI analytics platform, Multi-Sigma®, supports R&D through predictive modeling with limited datasets, Contribution Analysis, Multi-Objective Optimization, and AI-driven design of experiments.
We also introduced Multi-Sigma-Alchemy, which extends these capabilities to chemistry, materials science, and drug discovery. By generating descriptors from molecular structures, crystal structures, spectra, and images, the platform enables researchers to predict material properties and explore promising new compounds and materials.
Together, these capabilities demonstrate how AI can support a broad range of R&D activities, from process optimization to materials discovery.
Key Topics and Insights from Participants
The seminar prompted discussions on how AI could address challenges in research and industrial applications. Participants showed particular interest in experimental and process optimization, predictive modeling, scale-up, AI-driven design of experiments, and the integration of multiple AI models.
One recurring topic was the challenge of moving from laboratory experiments to production-scale processes. Researchers and engineers discussed how AI could help identify suitable operating conditions while accounting for differences in equipment, process behavior, and scale.
Another area of interest was AI Chain Analysis, which connects predictive models for individual process steps to evaluate and optimize overall process performance. Rather than treating each step independently, this approach makes it possible to examine how intermediate properties and operating conditions influence downstream results.
Participants also explored the potential of Bayesian optimization to identify promising experimental conditions with fewer trials, as well as Multi-Objective Optimization to balance competing performance requirements. Those working with computational fluid dynamics (CFD) and other computer-aided engineering (CAE) simulations were particularly interested in surrogate modeling as a way to reduce computational costs and combine simulation results with experimental data.
The discussions covered a broad range of applications, including CO2 capture and utilization, hydrogen production, catalyst development, chemical reactions and polymerization, battery materials, and functional polymers. We also received inquiries related to pharmaceuticals, bioprocesses, water treatment, food, and plant development.
In materials research, a common challenge was understanding how composition and processing conditions affect final material properties, and using those insights to determine which experiments to conduct next. These conversations highlighted the growing interest in data-driven approaches across process informatics, materials informatics, and simulation-based engineering.
Supporting the Practical Adoption of AI in R&D
Beyond providing AI analytics software, AIZOTH helps organizations apply AI to their specific research challenges through proof-of-concept (PoC) projects, data analysis support, AI implementation consulting, and custom R&D.
The SCEJ 57th Autumn Meeting gave us the opportunity to connect with researchers and engineers across academia and industry, gain a deeper understanding of their challenges, and discuss potential applications of AI in their work.
We would like to thank everyone who attended our presentation and luncheon seminar, participated in the discussions, or shared feedback through our survey. We look forward to continuing these conversations and helping researchers turn their data into actionable insights.
We have prepared various leaflets introducing analysis examples using Multi-Sigma®, the AI analytics platform developed by AIZOTH. These leaflets can be downloaded from the page below.
https://aizoth.com/en/service/multi-sigma/download/
For any questions or consultations regarding not only our products but also PoC projects, joint research, or research and development using AI and machine learning, please contact us using the inquiry form below.
