Refine
Document Type
- Conference Paper (4)
- Article (2)
Language
- English (6)
Is part of the Bibliography
- yes (6)
Institute
Review
- ja (5)
- peer-review (1)
Open Access
- ja (2)
Version
- published (6)
The Internet of Things (IoT) era facilitates new possibilities at the interface between the lifecycle of the product and the manufacturing disciplines allowing to add value and meet growing legal requirements. This work presents use cases from the automobile industry which are analyzed in order to derive requirements for data exchange. The analysis of the use cases shows clearly that we need to focus on the instance level for this purpose. This means that both, a specific unit of the product and the production system, must be considered. As a solution approach, a communication loop is proposed bringing the Digital Twin concept in a new field of usage. The key aspect of the proposed solution is the direct intercommunication between the Digital Twins of the product being produced as a Cyber-Physical System (CPS) and the production resources as Cyber-Physical Production Systems (CPPS). This solution enables the aggregation of production-related data within the products Digital Twin and therefore promotes maximal transparency throughout engineering disciplines, phases of the product lifecycle and the whole supply chain.
Adaptive production control in a modular assembly system based on partial look-ahead scheduling
(2019)
EU regulations on CO2 limits and the trend of individualization are pushing the automotive industry towards greater flexibility and robustness in production. One approach to address these challenges is modular production, where workstations are decoupled by automated guided vehicles, requiring new control concepts. Modular production control aims at throughput-optimal coordination of products, workstations, and vehicles. For this np-hard problem, conventional control approaches lack in computing efficiency, do not find optimal solutions, or are not generalizable. In contrast, Deep Reinforcement Learning offers powerful and generalizable algorithms, able to deal with varying environments and high complexity. One of these algorithms is Proximal Policy Optimization, which is used in this article to address modular production control. Experiments in several modular production control settings demonstrate stable, reliable, optimal, and generalizable learning behavior. The agent successfully adapts its strategies with respect to the given problem configuration. We explain how to get to this learning behavior, especially focusing on the agent’s action, state, and reward design.