Digital Twin for Pre-Deployment Validation of AI-Driven Safety-Critical Industrial Edge Control Loops
Authors: Sara Cavallero, Federico Tonini, Tony Chahoud, Giampaolo Cuozzo, Davide Borsatti, Walter Cerroni, Maurizio Fodrini, Riccardo Marini
Organizations: National Laboratory of Wireless Communications of CNIT (Wilab, CNIT), Italy · DEI, University of Bologna, Italy · WiLab CNIT, Italy · Fibercop
Industrial environments are increasingly characterized by the tight interaction among physical processes, communication infrastructures, and intelligent applications. In this context, Digital Twins (DTs) have emerged as a key technology for system analysis and optimization. However, existing DT solutions typically focus either on industrial processes or communication networks, while lacking an integrated and application-aware perspective. To fill this gap, this paper proposes a modular DT framework for industrial environments that jointly models physical processes, wireless communications, and application logic within a unified architecture. The feasibility of the proposed framework is experimentally validated through a real-world Proof-of-Concept (PoC) implemented in the BI-REX pilot line, involving a 5G-connected Autonomous Mobile Robot (AMR) transporting hazardous liquids and remotely controlled by an AI-driven application. The proposed DT is used to reproduce the behaviour of the real deployment and to investigate the impact of different placements of the AI application, including on-premise, edge, and remote cloud execution scenarios. Experimental results demonstrate a close agreement between DT predictions and PoC measurements in terms of both network-level metrics, such as Reference Signal Received Power (RSRP) and latency, and end- to-end application metrics, including application-level Round- Trip-Time (RTT). Moreover, the analysis shows how inaccuracies of network modeling can critically affect the feasibility of latency-sensitive industrial control loops, highlighting the potential of integrated DTs as tools for the pre-deployment design and validation of next-generation industrial systems.
Figures & tables
Fig. 1: Proposed Digital Twin architecture. The industrial scenario (process, assets, and network) and the DT domain (Process, Network, and Decision Modules) are bridged by a shared data infrastructure. This infrastructure enables a two-way flow: it pulls real-world measurements directly into the DT and, in turn, distributes the DT ’s services, including the synthetic data generated by its modules, to all external domains.
Fig. 2: Block diagram of the industrial PoC .
t
a(t)
gz(t)
p(t)
θ(t)
Label
1
a(1)
gz(1)
p(1)
θ(1)
Non-Fault
2
a(2)
gz(2)
p(2)
θ(2)
Non-Fault
⋮
⋮
⋮
⋮
⋮
Non-Fault
tf−A
a(tf−A)
gz(tf−A)
p(tf−A)
θ(tf−A)
Fault
tf−A+1
a(tf−A+1)
gz(tf−A+1)
p(tf−A+1)
θ(tf−A+1)
Fault
⋮
⋮
⋮
⋮
⋮
Fault
TABLE I: Example of one labeled time-series.
Fig. 3: Example of the Fault prediction using the trained CNN: the model anticipates the spill event from time series data.
Fig. 4: End-to-end communication and processing flow considered in the delay model. Blue arrows denote uplink sensor-data transmission, while red arrows represent downlink actuation commands.
Fig. 5: Real-world validation of the industrial PoC. The left image shows the outcome when the PoC is inactive (i.e., the water bottle falls due to a sudden obstruction), whereas the right image shows that the proposed solution prevents this event by accurately predicting the spill and activating the gripper in advance thanks to the high accuracy of the CNN model and the low latency provided by the OAI-based 5G network.
Fig. 6: Digital Twin Environment.
Symbol
Description
Value
TS
Motion data periodicity
100 ms
TP,UE
UE processing time
2 ms
TP,gNB
gNB processing time
4 ms
TCN
CN delay [ 21 ]
2 ms
TTCP
TCP overhead
7 ms
TT
CN-application server connection delay
1 ms
TABLE II: Input parameters setting.
PoC
DT
#1
#2
#1
#2
RT [ms]
66.16
65.24
69.12 ± 6.13
69.21 ± 6.32
TC [ms]
24.59
24.77
24.30 ± 0.80
24.25 ± 0.78
PS [dBm]
-77
-79
-79 ± 0.04
-81 ± 0.08
TABLE III: Comparison of the KPIs between the PoC and the DT for two blockage events. For the DT , results are reported as mean ± standard deviation over NE=50 , while NE=10 for the PoC .
Fig. 7: Locations of the two blockage positions used for both PoC experiments and DT evaluations. Blockage #1 is placed near the gNB , while Blockage #2 is farther.
Fig. 8: Effects of different margins ( M ) and application requirements ( TI ) on the time budget for TT . Reference delay values for different cloud solutions are reported as horizontal lines.
Digital twins have emerged as a foundational technology within the context of Industry 4.0, offering a paradigm for the real-time virtual representation of physical systems. However, managing their growing complexity, particularly in distributed industrial environments, requires intelligent architectures capable of autonomous decision-making, dynamic adaptability, and inter-agent coordination. This systematic review explores the intersection between Multi-Agent Systems and Digital Twins, with a particular focus on predictive maintenance applications in resource-constrained contexts. Through a critical analysis of over 547 papers published in high-impact journals (IEEE Transactions, Nature, Elsevier, MDPI), we establish a taxonomy of existing hybrid architectures, identify persistent technological bottlenecks, and formulate three open research questions concerning: (i) the deployment of artificial intelligence on resource-constrained microcontrollers, (ii) distributed multi-node coordination via lightweight communication protocols, and (iii) the hierarchical orchestration of Digital Twins toward smart factory control integrating residual life estimation and explainable Artificial Intelligence. The results of this analysis reveal that, despite significant progress, no existing system offers an integrated embedded-distributed hierarchical solution that simultaneously meets the requirements of Industry 5.0.
Korota Arsène Coulibaly, Mohamed Hamlich
LCCPS Lab, ENSAM, Hassan II University of Casablanca, 150 Bd du Nil, Casablanca, 20670,, Morocco
Industrial automation is being transformed by digitalization and the increasing use of cyber-physical systems. Modern production environments require greater adaptability, faster reconfiguration, and more intuitive human-machine interaction. However, traditional rule-based systems rely on fixed logic and cannot autonomously adapt to changing conditions. Consequently, current automation systems lack a systematic approach for integrating adaptive and generalizable reasoning capabilities for interpreting, planning, and executing user tasks across dynamic environments and heterogeneous components. This dissertation proposes a three-layer framework that integrates large language models (LLMs), digital twins, and automation systems into an autonomous system. Autonomy is defined as a design property assigned to system components and enabled through LLM-based reasoning to achieve adaptive, goal-oriented behavior. The Task-Process-Service-Resource (TPSR) model is introduced to transform user tasks into executable processes. Four LLM roles are identified: process orchestration, service matching, digital resource generation, and agent-as-a-service. Five peer-reviewed studies develop and refine these concepts using the design science research methodology. Case studies and prototypes demonstrate adaptive task planning, event-driven control, simulation-based parameterization, and digital model generation. Results show high task executability, command correctness, and content-generation accuracy while reducing manual effort. The framework enables the integration of LLM-based reasoning into industrial automation systems and improves adaptability and usability. Limitations include dependence on accurate digital representations, the computational demands of LLMs, and the need for human intervention in safety-critical situations.
Yuchen Xia
Institut für Automatisierungs- und Softwaresysteme (IAS) der Universität Stuttgart
Network Digital Twins (NDTs) enable safe what-if analysis for 6G cloud-edge infrastructures, but adoption is often limited by fragmented workflows from telemetry to validation. We present a data-driven NDT framework that extends 6G-TWIN with a scalable pipeline for cloud-edge telemetry aggregation and semantic alignment into unified data models. Our contributions include: (i) scalable cloud-edge telemetry collection, (ii) regime-aware feature engineering capturing the network's scaling behavior, and (iii) a validation methodology based on Sign Agreement and Directional Sensitivity. Evaluated on a Kubernetes-managed cluster, the framework extrapolates performance to unseen high-load regimes. Results show both Deep Neural Network (DNN) and XGBoost achieve high regression accuracy (R2 > 0.99), while the XGBoost model delivers superior directional reliability (Sa > 0.90), making the NDT a trustworthy tool for proactive resource scaling in out-of-distribution scenarios.