Economic and flexible decentral self-optimizing production (SOPROD): final report

Steel is produced via a series of different, interdependent processes. In the past, most of these processes have been individually optimized, but rarely considered constraints of follow-up aggregates. A huge improvement potential can therefore be achieved by optimizing the process chain as such, introducing smart processes and smart products that autonomously interact with each other, which is the main idea of the SOPROD project. The project demonstrates the implementation self-optimization, a specific notion of self-organization, by using agent-based digital twins to allow products telling the machines the best setup for their production. Digital twins, built upon the concept of autonomous, interacting and communicating agents, are deployed to represent steel products and machinery. They carry mathematical models to predict their future states. With this approach, a steel product - respectively the associated digital twin - knows its own temperature now and in the next hours, as well, as it can predict the amount of scale on the surface. With this principally new information from the models, products can also anticipate an optimized production setup and inform the next process step about this optimum. The SOPROD project validated its concepts and developments at two dedicated real-world use cases and showed, that the idea of self-organization is indeed capable of improving steel production chains.