Ablaufplanung mit alternativen Prozessplänen

Watermeyer, Kai

doi:10.1007/978-3-658-12093-1

Cited by 5 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…-Job release management (JReM): Release jobs of different product types into the production system at certain points in time out of a given job pool. A more formal description of the control context orients itself on a scheduling formulation by Watermeyer (2016) and is as follows: A set of n J jobs J is processed by a set of n M modular workstations M and transported by n V AGVs V with speed v V . Each workstation M k comprises an Inboundand an Outbound-Buffer (IB and OB) with equal capacity n bu f utilizing First In -First Out (FIFO).…”

Section: Modular Production Control Problem Formulationmentioning

confidence: 99%

Modular production control using deep reinforcement learning: proximal policy optimization

2021

View full text Add to dashboard Cite

EU regulations on $$\textit{CO}_2$$ CO 2 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.

show abstract

Section: Modular Production Control Problem Formulationmentioning

confidence: 99%