mlsanityData to 14 Oct 2026

Balancing Specialization and Centralization: A Multi-Agent Reinforcement Learning Benchmark for Sequential Industrial Control

arXiv 2510.20408 ↗First posted 23 Oct 2025cs.LG cs.AI cs.MAcited by 0

Not on the map. Why?

Abstract (arXiv)

Autonomous control of multi-stage industrial processes requires both local specialization and global coordination. Reinforcement learning (RL) offers a promising approach, but its industrial adoption remains limited due to challenges such as reward design, modularity, and action space management. Many academic benchmarks differ markedly from industrial control problems, limiting their transferability to real-world applications. This study introduces an enhanced industry-inspired benchmark environment that combines tasks from two existing benchmarks, SortingEnv and ContainerGym, into a sequential recycling scenario with sorting and pressing operations. We evaluate two control strategies: a modular architecture with specialized agents and a monolithic agent governing the full system, while also analyzing the impact of action masking. Our experiments show that without action masking, agents struggle to learn effective policies, with the modular architecture performing better. When action masking is applied, both architectures improve substantially, and the performance gap narrows considerably. These results highlight the decisive role of action space constraints and suggest that the advantages of specialization diminish as action complexity is reduced. The proposed benchmark thus provides a valuable testbed for exploring practical and robust multi-agent RL solutions in industrial automation, while contributing to the ongoing debate on centralization versus specialization.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 10 of its 25 references that are in the map. First posted on arXiv 23 Oct 2025.

Builds on

10 of 25 references.

References · 10

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  3. 2019The StarCraft Multi-Agent Challengecited by 597
  4. 2020Learning to Utilize Shaping Rewards: A New Approach of Reward Shapingcited by 114
  5. 2021Scalar reward is not enough: A response to Silver, Singh, Precup and Sutton (2021)cited by 55
  6. 2024Comprehensive Overview of Reward Engineering and Shaping in Advancing Reinforcement Learning Applicationscited by 38
  7. 2024Excluding the Irrelevant: Focusing Reinforcement Learning through Continuous Action Maskingcited by 19
  8. 2023ContainerGym: A Real-World Reinforcement Learning Benchmark for Resource Allocationcited by 6
  9. 2025SortingEnv: An Extendable RL-Environment for an Industrial Sorting Processcited by 3
  10. 2025Leveraging Genetic Algorithms for Efficient Demonstration Generation in Real-World Reinforcement Learning Environmentscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Evolutionary Warm-Starts for Reinforcement Learning in Industrial Continuous Control
  2. 2020The reinforcement learning-based multi-agent cooperative approach for the adaptive speed regulation on a metallurgical pickling line ↗
  3. 2026Learning Multi-Agent Task Assignment and Navigation in the Factory: from Simulation to Real Robots
  4. 2026Event-Driven Reinforcement Learning Enables Long-Horizon Control in Semiconductor Fabrication
  5. 2026A MARL Centered Reference Architecture for Large Language Model Augmentation in Smart Manufacturing
  6. 2025A Survey of Reinforcement Learning for Optimization in Automation
  7. 2025SHaRe-RL: Structured, Interactive Reinforcement Learning for Contact-Rich Industrial Assembly Tasks
  8. 2026Augmenting PID Control with Deep Reinforcement Learning: A Hybrid Approach to the Industrial Benchmark