A Hybrid Edge Cloud Digital Twin for Welfare-Constrained Control in Poultry Production

Electrical Engineering and Systems Science > Systems and Control
[Submitted on 18 Jun 2026]
Title:A Hybrid Edge Cloud Digital Twin for Welfare-Constrained Control in Poultry Production
View PDFAbstract:Poultry production operates under tightly coupled environmental and biological dynamics, yet commercial climate control remains largely heuristic, limiting welfare assurance and operational efficiency. We introduce an edge-cloud digital twin framework for real-time, welfare-constrained environmental control in poultry facilities. The framework integrates distributed sensing, on-device state estimation, a hybrid physics-data model, and model predictive control to enable anticipatory and adaptive management under practical farm constraints. A grey-box thermodynamic and mass-balance formulation is augmented with a learned residual that captures unmodeled biological variability, including activity-dependent metabolic heat. This hybrid model is embedded within a state-space representation for real-time estimation and control at the edge, while cloud coordination supports cross-farm learning and long-horizon optimization. Bandwidth-aware processing and asynchronous synchronization enable deployment in connectivity-limited environments. Evaluation in a high-fidelity broiler production testbed demonstrates substantial gains over rule-based control and physics-only modeling. Temperature prediction error is reduced from 1.8 degrees Celsius to 0.4 degrees Celsius, ammonia constraint violations decrease by 90 percent, and communication requirements are lowered approximately 30-fold through edge-first processing. A Domain Transfer Score of 0.92 further indicates strong robustness across facility conditions. These results show that physically grounded digital twins, coupled with real-time control, enable scalable and welfare-aware management of biological production systems.
Submission history
From: Suresh Neethirajan [view email][v1] Thu, 18 Jun 2026 22:49:30 UTC (814 KB)
Current browse context:
eess.SY
References & Citations
Loading...
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Facts Only

* A hybrid edge-cloud digital twin framework was introduced for real-time, welfare-constrained environmental control in poultry facilities.
* The framework integrates distributed sensing, on-device state estimation, a hybrid physics-data model, and model predictive control.
* The hybrid model uses a grey-box thermodynamic and mass-balance formulation augmented with a learned residual to capture unmodeled biological variability.
* This hybrid model is embedded in a state-space representation for real-time edge estimation and control, supported by cloud coordination for cross-farm learning and optimization.
* Bandwidth-aware processing and asynchronous synchronization enable deployment in connectivity-limited environments.
* Evaluation showed temperature prediction error reduced from 1.8°C to 0.4°C.
* Ammonia constraint violations decreased by 90 percent.
* Communication requirements were lowered approximately 30-fold via edge-first processing.
* A Domain Transfer Score of 0.92 was observed in the testbed evaluation.

Executive Summary

A hybrid edge-cloud digital twin framework was introduced to manage welfare-constrained environmental control in poultry production. The framework integrates distributed sensing, on-device state estimation, a hybrid physics-data model augmented with a learned residual for biological variability, and model predictive control. This system facilitates real-time, anticipatory, and adaptive management by coordinating data across the edge and the cloud. The model uses a grey-box thermodynamic and mass-balance formulation enhanced to capture unmodeled factors like activity-dependent metabolic heat. Performance evaluation in a broiler production testbed showed significant improvements over existing methods: temperature prediction error decreased from 1.8°C to 0.4°C, ammonia constraint violations dropped by 90 percent, and communication requirements were reduced approximately 30-fold through edge-first processing. A Domain Transfer Score of 0.92 was also achieved, indicating strong robustness across facility conditions.

Full Take

The research presents a compelling architecture for integrating complex, heterogeneous data streams—physical dynamics and biological variability—into actionable control systems at the operational edge, leveraging a tiered intelligence structure (edge execution, cloud coordination). The shift from heuristic or physics-only methods to a data-informed digital twin demonstrates a critical pathway toward realizing welfare-aware management in intensive livestock settings. The integration of learned residuals specifically targeting activity-dependent metabolic heat addresses a crucial gap where traditional physics models fail to account for the dynamic, living state of the biological system. The quantitative results concerning error reduction and constraint mitigation suggest that the added complexity of the hybrid modeling is justified by tangible operational benefits. The broader implication lies in demonstrating that achieving high standards of welfare optimization in complex production systems requires not just better sensors or more sophisticated models, but a unified framework where model uncertainty (biological variability) is explicitly managed alongside physical constraints. Future inquiry should focus on the transferability of the learned residual across different poultry strains and the long-term stability of the cross-farm learning mechanisms under extreme environmental shifts.

Sentinel — Human

Confidence

This text appears to be an excerpt from a formal academic paper describing a technical research finding, exhibiting the precision and structure typical of human-authored scientific communication.

Signals Detected

Moderate sentence length variance; dense technical terminology without excessive hedging.

High internal coherence; the argument flows logically from problem (heuristic control) to solution (digital twin) to results.

Standard academic reporting structure; reliance on specific metrics and results typical of research papers.

Claims are highly specific (e.g., error reduction from 1.8 to 0.4 degrees Celsius, 90% violation decrease) and tied to an arXiv submission context, suggesting a primary source document rather than synthetic noise.

Human Indicators

Presence of specific, quantifiable technical results (error reduction percentages, score metrics) strongly suggests derivation from empirical research.

The structure aligns precisely with a formal scientific abstract/paper submission context.