As part of the research project FENI-X, USU is contributing its AI expertise in the field of IT and service management as an industry partner. The aim of the project is to strengthen the resilience of global production networks through the targeted use of artificial intelligence and to make manufacturing companies more resistant to disruptions and uncertainties. FENI-X is funded by the German Federal Ministry of Education and Research (BMBF). The initiator is the Karlsruhe Institute of Technology (KIT), and the project is managed by the Project Administration Karlsruhe (PTKA).
New Methodology for Increasing Resilience
The FENI-X methodology is based on a modular resilience framework. This framework defines resilience metrics, evaluates the status quo of companies and derives measures to strengthen their resilience. The measures are tested in four use cases that cover different planning levels and examine specific resilience-enhancing mechanisms. USU is actively involved in two of these use cases, contributing its expertise as well as its software solutions in the areas of orchestrating complex IT landscapes and risk management using digital workflows. Finally, the scientific findings and best practices will be made available to a broad industrial and academic audience.
Strong Positioning of USU in the Project Consortium
As an industry partner, USU is well positioned within the project thanks to its AI expertise and is contributing a clear exploitation plan to ensure that the project results can be transferred into market-ready solutions in the future.
"Through FENI-X, we combine modern AI methods with our proven expertise in IT and service management. Our goal is for IT systems in the future to provide concrete recommendations for actions that go beyond traditional anomaly detection and associated warning signals. To achieve this, our software solutions collect machine data, analyze it and identify deviations or errors. Subsequently, recommendations for action are generated using a decision model based on generative AI. In practice, for example, if a production machine stops due to an error, the LLM compares the information with the knowledge database to determine the type of error and evaluates possible courses of action—such as whether replacing the machine is necessary or whether process adjustments are sufficient. Machine operators and responsible staff therefore receive a solution proposal and are actively supported in resolving the issue. This sustainably strengthens the resilience of industrial systems," explains Henrik Oppermann, Director Research at USU.
Current Project Status and Initial Results
The FENI-X research project is scheduled to run for a total of around three years, of which nearly one year has already been successfully completed. The consortium is currently engaged in detailed project planning and developing the concepts and measures to be investigated in the use cases. USU has implemented its first technical prototype, which will now be specifically expanded in the second project year to include algorithms for anomaly detection as well as a decision-making model.
New Methodology for Increasing Resilience
The FENI-X methodology is based on a modular resilience framework. This framework defines resilience metrics, evaluates the status quo of companies and derives measures to strengthen their resilience. The measures are tested in four use cases that cover different planning levels and examine specific resilience-enhancing mechanisms. USU is actively involved in two of these use cases, contributing its expertise as well as its software solutions in the areas of orchestrating complex IT landscapes and risk management using digital workflows. Finally, the scientific findings and best practices will be made available to a broad industrial and academic audience.
Strong Positioning of USU in the Project Consortium
As an industry partner, USU is well positioned within the project thanks to its AI expertise and is contributing a clear exploitation plan to ensure that the project results can be transferred into market-ready solutions in the future.
"Through FENI-X, we combine modern AI methods with our proven expertise in IT and service management. Our goal is for IT systems in the future to provide concrete recommendations for actions that go beyond traditional anomaly detection and associated warning signals. To achieve this, our software solutions collect machine data, analyze it and identify deviations or errors. Subsequently, recommendations for action are generated using a decision model based on generative AI. In practice, for example, if a production machine stops due to an error, the LLM compares the information with the knowledge database to determine the type of error and evaluates possible courses of action—such as whether replacing the machine is necessary or whether process adjustments are sufficient. Machine operators and responsible staff therefore receive a solution proposal and are actively supported in resolving the issue. This sustainably strengthens the resilience of industrial systems," explains Henrik Oppermann, Director Research at USU.
Current Project Status and Initial Results
The FENI-X research project is scheduled to run for a total of around three years, of which nearly one year has already been successfully completed. The consortium is currently engaged in detailed project planning and developing the concepts and measures to be investigated in the use cases. USU has implemented its first technical prototype, which will now be specifically expanded in the second project year to include algorithms for anomaly detection as well as a decision-making model.
