EUVIMEDEUVIMED entdecken
Crossref

Supervisory Control of Automated Manufacturing Systems: A Unified Survey of Petri Net Models From Bounded to Unbounded Frameworks

Hamidu Mamudu, Hesuan Hu

IEEE Transactions on Automation Science and Engineering · 2026 · 23 · 10587-10606

DOI: 10.1109/tase.2026.3699271 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Automation Science and Engineering, 2026. Verfasst von Hamidu Mamudu und 1 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities, New Century Excellent Talents in University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The transition from classical automated manufacturing to high-volume, asynchronous System-of-Systems (SoS) architectures has exposed a fundamental tractability collapse in Petri-net-based supervisory control. Existing frameworks face an inherent expressiveness–control trade-off: bounded models guarantee decidability only by imposing artificial material-flow constraints, while unbounded Petri nets (UPNs) capture manufacturing realism but suffer from state-space explosion and observational delay. This paper formalizes eight structural shortcomings (S1–S8), referred to as the Detectability Fight, which prevent existing paradigms from achieving simultaneous scalability and semantic fidelity. We show that supervisory approaches based on non-bisimilar abstractions cannot ensure safety and liveness without sacrificing performance. To resolve this limitation, we introduce the Transition-bounded Unbounded Petri Net (tUPN) framework. By shifting the supervisory locus from place-centric capacity constraints (stock) to transition-centric execution causality (flow), tUPN enables deterministic supervisory control with linear-time complexity $O(n)$ , even in the presence of unbounded work-in-process. We further demonstrate that the tUPN constructs $(\mathcal {K},\kappa _{t},\phi _{t})$ admit direct realization within ISA-95 manufacturing hierarchies and IEC 61131-3-compliant PLC runtime environments, providing a formally verified supervisory envelope for safe AI-assisted optimization. This work establishes an execution-centric foundation for scalable, plug-and-play supervisory control in next-generation manufacturing systems.Note to Practitioners—This work is motivated by the computational growth observed when traditional supervisory control methods are applied to large-scale, high-throughput manufacturing systems. In practice, engineers are often forced to impose artificial buffer limits simply to keep control logic computationally feasible, thereby distorting physical plant behavior. The tUPN framework addresses this issue by regulating the execution of actions rather than the accumulation of material. Control is applied through transition-level execution bounds instead of buffer counting. For practitioners, this yields three practical outcomes: 1) real-time scalability, as supervisory decisions scale linearly with the number of machines rather than work-in-process volume; and 2) standards-aligned implementation, with supervisory logic directly realizable using IEC 61131-3 PLC constructs and ISA-95 production policies; and 3) safe integration of AI optimizers, which may adjust operational targets while remaining constrained by a formally verified supervisory envelope. As a result, tUPN supports deployment of plug-and-play manufacturing cells that retain formal correctness without sacrificing industrial scalability.

Förderung

  • National Natural Science Foundation of China
  • Fundamental Research Funds for the Central Universities
  • New Century Excellent Talents in University
  • Academic Research Fund Tier 1 by Ministry of Education in Singapore
  • Academic Research Fund Tier 2 by Ministry of Education in Singapore
  • Major Fundamental Research Program of the Natural Science Foundation of Shaanxi Province

Zitieren

APA

Mamudu, H., & Hu, H. (2026). Supervisory Control of Automated Manufacturing Systems: A Unified Survey of Petri Net Models From Bounded to Unbounded Frameworks. IEEE Transactions on Automation Science and Engineering, 23, 10587-10606. https://doi.org/10.1109/tase.2026.3699271

Vancouver

Mamudu H, Hu H. Supervisory Control of Automated Manufacturing Systems: A Unified Survey of Petri Net Models From Bounded to Unbounded Frameworks. IEEE Transactions on Automation Science and Engineering. 2026;23:10587-10606. doi:10.1109/tase.2026.3699271

BibTeX
@article{Mamudu2026,
  author = {Mamudu, Hamidu and Hu, Hesuan},
  title = {Supervisory Control of Automated Manufacturing Systems: A Unified Survey of Petri Net Models From Bounded to Unbounded Frameworks},
  journal = {IEEE Transactions on Automation Science and Engineering},
  year = {2026},
  volume = {23},
  pages = {10587--10606},
  doi = {10.1109/tase.2026.3699271}
}
Kostenlos starten →