Description
Dynamic scheduling and planning for MaaS are made possible by exploiting formal models and forming new scheduling/control algorithms from Hybrid Systems theory. It is responsible for the generation of feasible and optimized operational plans, utilizing as input the formal models, AI-based predictions and optimization objectives. Regarding the optimization formulations, the component supports algorithmic solutions for relevant optimization problems of the manufacturing and supply chain domains, utilizing the UniMaaS Manufacturing Service Chain (MSC) abstraction, specifically, scheduling, task sequencing, routing, flow network optimization, resource allocation, and dynamic reconfiguration of MSCs in response to operational changes. Depending on the manufacturing setting and the optimization objectives, different algorithmic approaches may be employed, ranging from combinatorial optimization and Operations Research techniques to Model Predictive Control (MPC), Multi-Criteria Decision-Making (MCDM), Reinforcement Learning (RL), and hybrid optimization methods.
Main Innovations
- Dynamic Scheduling and Planning for MaaS
- Fast reconfiguration leveraging current state of the infrastructure
- Incorporate AI predictions and Circularity aspects in the planning algorithms
- Optimize the supply chain over the horizon
Licence
MIT
TRL
3 → 6
