Driving production performance in North America through AI and reaction kinetics

Born and raised in China, Jiakang Chen traveled to the United States as part of an exchange program. After completing his Ph.D. in Chemical Engineering at the University of Houston, he joined BASF in 2023. Since then, he has been helping production plants increase yields and implement leaner, more efficient processes. Learn more about Jiakang’s life and work below.

Where did your interest in science come from?

I am the first person in my family to choose a technology related major, and my path into science really started with curiosity. In middle school, I remember standing in a chemistry lab during a simple experiment where iron reacted with hydrochloric acid to form bubbles. When I saw that, I felt like I was actually creating chemistry myself. That moment stayed with me, and I kept thinking about how the reaction worked and what might happen if we changed different conditions. It was no longer just a concept from a textbook; it felt real, alive, something I could explore. From that point on, I set a goal for myself to work in science and be part of discovering and creating things like that.

When did you join BASF?

I came to the U.S. from China in 2017, as part of an exchange program between my home university, the East China University of Science and Technology, and the University of Houston. The idea was to learn what the course looked like in the U.S. and the study environment. Studying in this new educational environment pushed me to become more independent and proactive.

I also had the opportunity to join a research lab and that experience had a significant impact on me. I always knew I wanted to go into the industry, but the experience pushed me to pursue research more extensively, so I chose to pursue a Ph.D.

I had known about BASF for a long time. As engineers, we often study many of the processes that BASF has pioneered, and it has always been my dream to be a part of creating new innovations. When I was in graduate school, about to get my doctorate, I actively reached out to R&D staff via LinkedIn not knowing that one day I would be working alongside them. Incidentally, there was a role in the reaction engineering team available right as I was about to graduate, and I felt really fortunate when I got selected for the position. I have been with the company for over three years now and I enjoy working here every single day.

Tell me about your current position at the company.

I am currently a research engineer, and what I really value is the level of ownership and responsibility I’m able to take. When I was in graduate school, I never imagined I would have the opportunity to work so closely with real plant operations. However, right from my first month at work, I was able to support plant operations directly through troubleshooting, optimization, and performance improvement, while also contributing to more fundamental research. I enjoy being in that space where I can connect practical challenges with deeper scientific understanding.

Our team supports more than 40 plants across North America. I find it really exciting to work across different businesses and processes, because every problem is different and pushes me to think in new ways. This exposure has helped me develop a much broader, system-level understanding of chemical manufacturing.

We also actively collaborate with external partners such as universities and national laboratories. I’m always interested in new ideas and emerging technologies, and these collaborations give me the chance to explore cutting-edge concepts at a deeper level and think about how we can apply them in our industry.

I believe that continuous learning is very important to keep developing as a technologist. Before joining BASF, my work was mainly focused on reaction engineering and kinetic modeling. Over time, I developed a strong interest in coding, which led me to pursue a degree in computer science. Now, I’m really interested in integrating computer science and AI into chemical engineering and using data-driven approaches to improve process performance and accelerate the development of new chemistry.

How do you use artificial intelligence (AI) in your work?

I’m really excited about using AI to solve real problems in the chemical industry, not just as a tool, but as a way to rethink how we approach complex systems. In my work, I focus on two main directions.

First, I work closely with plant teams to bring AI directly into real operations. The goal is not just to analyze data, but to combine plant data with domain knowledge, engineering intuition, and large language models (LLMs) to create smarter and more practical solutions. What excites me most is grounding these models in real plant behavior, so they can reflect how systems actually work, not just how we assume they work. This opens new ways to understand complex kinetics and interactions, and to make more reliable and predictive decisions in day-to-day operations.

Second, I collaborate with our colleagues in catalyst research as well as external partners on applying AI to catalyst design. Traditionally, this process relies heavily on experience and trial and error. I’m interested in changing that by combining AI with high throughput experimentation, so we can explore a much larger design space and at the same time, do this much more rapidly. Instead of just testing ideas, I hope to use AI to actively suggest better candidates, guided by both theory and experiments. To me, this is where AI really becomes powerful, helping us not only understand chemistry better, but also create better materials more efficiently.

Tell me about the project you recently worked on at BASF’s Hannibal, Missouri site.

At the Hannibal site, we have been working closely with the Agricultural Solutions team for quite a long time, so we had already built a strong understanding of their process and system. Through this collaboration, we identified an opportunity to further improve overall plant efficiency.

A kinetic model describes how fast chemical reactions take place and how those reaction rates change under varying operating conditions, such as temperature or feed rate. In Hannibal, we used a kinetic model to evaluate the manufacturing process and the reaction pathways leading to both the desired products and byproducts. When developed correctly, the model would identify the reaction conditions that would optimize raw material utilization and overall efficiency.

Kinetic modeling can capture both the process dynamic and impact of equipment or infrastructure. In this case, after designing experiments with the plant's operations team and ensuring data accuracy, the kinetic model provided recommendations for process optimizations.

I traveled to the site and implemented trial plans with the operations team once the model was finalized. At BASF, we never compromise on safety, and a small-scale plan trial ensures we are able to produce the same quality product in a safe way. Only after the trial was completed successfully, meaning safely and to our high-quality standards, will the manufacturing team apply it to the entire manufacturing process.

In this case, after a year of data analysis, model development, and pre-work, we were able to successfully complete the trial in about a week. Following the successful trial, the manufacturing team will continue to use this new process, and I will move onto the next project at one of our other manufacturing sites in North America.

For questions, please contact  Mariana Licio.

For media inquiries, please contact Lisa Brown.