KEEN Research Project Successfully Finished
The publicly funded project KEEN (AI Incubator Labs in the Process Industry) started in early 2020 with the mission to explore and leverage methods of artificial intelligence (AI) in the process industries. Now, three and a half years later, the project has successfully finished.
The project was part of the Geman Federal Ministry for Economic Affairs and Climate Action´s (BMWK) program Digital Technologies, and in particular of the Innovation Competition Artificial Intelligence. The 26 funded projects in this competition covered a broad range of industrial and societal fields, but KEEN was the only project that focused on the chemical and pharmaceutical industries, the third largest industrial sector in Germany.
Developing new tools and methods for industrial practice
Within KEEN, INOSIM has collaborated with major industrial partners and universities to work on a variety of new methods and tools for the digital support of industrial batch plants. With Merck and TU Dortmund, INOSIM developed a software tool for the plant-wide and unit-specific predictive decision support in industrial batch plants. With Bayer, ABB, and TU Dresden, INOSIM contributed to the development of new methods for the automatic detection of batch phases from historic plant data.
Harnessing AI To Create Optimized Production Schedules
Last but not least, INOSIM collaborated with Bayer to evaluate and assess the use of AI-based methods for the reactive real-time scheduling of batch plants. Short-term access to optimal production schedules, for example following spontaneous changes, offers numerous advantages in terms of robustness, economy, and ultimately customer satisfaction, as delays are minimized. Deep Reinforcement Learning (DRL) for optimized planning in a typical batch production plant of the chemical industry was tested and evaluated here for the first time. The pilot study shows how DRL can be implemented with an approach based on discrete-event simulation with INOSIM. It is expected that the application of DRL in the chemical industry is a promising research and innovation direction and that DRL can complement established methods such as process simulation and mathematical programming. Due to its large promise, INOSIM plans to continue this work after the conclusion of the KEEN project.
Many of the promising results of the KEEN project have been published in a special issue of Chemie Ingenieur Technik (volume 95, issue 7). That includes a publication describing the results of the work of INOSIM and Bayer on AI-based reactive scheduling.
General Inquiries 
