Human-centered Artificial Intelligence in the Chemical Industry

Project Goal

Human-AI partnership for the chemical industry

The chemical industry is facing a major challenge. The large data volumes of the complex logistical processes are now too extensive to be fully utilized by employees. Complete automation of data processing, however, would be too inflexible.

The solution is for humans and AIs to work together. Here, AIs can be used well in processing the data streams, and humans can review the processes and evaluate them depending on the possible consequences.

The development of a modular system from which individual solutions can be developed in companies is the goal of this project. Employees will be supported in identifying process problems at an early stage and selecting suitable solutions.

Further information on hKI-Chemie can also be found here

Our Contribution

To achieve this goal, it is necessary that humans can comprehend the decision-making mechanisms of the AI systems and, if necessary, take corrective action.
Explainability and interactivity are the basis for a functioning cooperation and the focus of our research in the project.
In this context, the project group Comprehensible Artificial Intelligence develops algorithms that are evaluated in close cooperation with our project partners.

Project Partner

  • Universität Duisburg-Essen (Department of Distributed Systems, Department of Psychological Research Methods)
  • Fraunhofer IIS (Projektgruppe Comprehensible Artificial Intelligence)
  • Continental Automotive GmbH
  • Evonik Industries AG
  • InfraServ Wiesbaden Technik GmbH & Co. KG