Design Of a Plug-And-Produce, Resilient Manufacturing Control System Using Asset Administration Shell: Based on the MADE REACT project
Translated title
Design Of a Plug-And-Produce, Resilient Manufacturing Control System Using Asset Administration Shell
Authors
Kristensen, Jeppe ; Bonde, Lucas Nesgaard ; Stensgaard, Silas Porsgaard
Term
4. term
Education
Publication year
2026
Submitted on
2026-05-28
Abstract
This report examines how to design and implement a plug-and-produce manufacturing control system that makes factories more flexible and resilient to change. The project is inspired by the MADE REACT initiative and focuses on enabling flexible production through standardized communication and digital representations of production equipment. The report introduces key concepts in modular manufacturing, where the production line is made up of smaller, interchangeable modules, and interoperability, meaning that different machines and IT systems can work together smoothly. It also discusses what characterizes robust and resilient manufacturing systems that can adapt quickly to disruptions or new requirements. A particular focus is placed on the use of Asset Administration Shells, which are digital information models of physical devices, MQTT-based communication, which allows devices to exchange data in a loosely coupled way, and PackML-based orchestration, which provides a standardized way to control machine states and sequences in production. Based on these technologies, a control architecture is designed and implemented to demonstrate how standardized information models and a flexible communication setup can support dynamic integration and flexible orchestration of manufacturing resources. The implemented solution is evaluated in terms of interoperability between systems, the ability to change and extend the system (reconfigurability), and the overall robustness and adaptability (resilience) of the manufacturing system. The results indicate that the developed solution can support more adaptive and flexible manufacturing systems. At the same time, the report highlights challenges, especially regarding standardization, orchestration complexity, and deployment at large industrial scale.
Denne rapport undersøger, hvordan man kan designe og implementere et plug-and-produce styringssystem til produktion, der gør fabrikker mere fleksible og modstandsdygtige over for ændringer. Projektet er inspireret af MADE REACT-initiativet og har fokus på at skabe fleksibel produktion ved hjælp af standardiseret kommunikation og digitale beskrivelser af produktionsudstyr. Rapporten gennemgår centrale begreber inden for modulær produktion, hvor produktionslinjen består af mindre, udskiftelige moduler, samt interoperabilitet, dvs. at forskellige maskiner og systemer kan arbejde problemfrit sammen. Derudover behandles, hvad der kendetegner robuste og resiliente produktionssystemer, som kan omstilles hurtigt ved forstyrrelser eller nye krav. Et særligt fokus er brugen af Asset Administration Shells, som er digitale informationsmodeller af fysiske enheder, MQTT-baseret kommunikation, der gør det muligt for enheder at udveksle data løst koblet, samt PackML-baseret orkestrering, som er en standardiseret måde at styre produktionsmaskiners tilstande og sekvenser på. På baggrund af disse teknologier bliver der designet og implementeret en styrearkitektur, der viser, hvordan standardiserede informationsmodeller og en fleksibel kommunikationsstruktur kan understøtte dynamisk integration og fleksibel styring af produktionsressourcer. Løsningen bliver vurderet med hensyn til, hvor godt den understøtter samarbejde mellem systemer (interoperabilitet), muligheden for at ændre og udvide systemet (rekonfigurerbarhed) og systemets robusthed og tilpasningsevne (resiliens). Resultaterne peger på, at den udviklede løsning kan bidrage til mere adaptive og fleksible produktionssystemer. Samtidig fremhæver rapporten udfordringer, særligt i forhold til ensartede standarder, kompleksiteten i orkestreringen og implementering i stor industriel skala.
[This abstract has been rewritten with the help of AI based on the project's original abstract]
Keywords