COGNIPLANT project’s approach consists in a hierarchical advanced monitoring and supervisory control that give a comprehensive vision of the plants’ production performance as well as the energy and resource consumption. Data collected from the production plants’ equipment and sensors will be structured in a data virtualization layer. Advanced data analytics will be applied to extract valuable information (e.g. patterns, indicators) from the processes and their effect on the production plant’s overall performance. An innovative Digital twin model of the factories will be created. Based on the conclusions extracted from the advanced data analysis, optimal operation plans will be designed and simulated in the digital twin models, coupled with a real time reactive scheduling tool. Finally, the consequences of these operations will be compared with the patterns extracted from the advanced data analysis, to measure the effectiveness of the prescriptive operations. The COGNIPLANT solution will consist of three main layers:
- Sensing and Data Virtualisation level ► “Co-Digitise”: collect and structure the data from the different sensors and equipment for its further analysis.
- Advanced data analytics level ► “Co-Analyse”: data processing, application of advanced methods of process mining, big data, data mining, etc.
- Virtual Model and simulation level ►“Co-Decide”: digital twin, decision making, generation of operational plans, prescriptive general and edge processing.