digital twins

Rising complexity in industrial systems undermines the reliability and safety assurance capabilities of conventional maintenance paradigms. Method: This study proposes a digital twin–driven predictive maintenance (PdM) framework, featuring a hierarchical technical architecture and an AI-empowered industrial application taxonomy. It integrates IoT sensing, machine learning, and real-time big data analytics to enable sensor-driven adaptive virtual twin modeling and dynamic performance optimization. Contribution: First, it systematically traces the evolutionary trajectory of digital twins in PdM—from static one-to-one mapping to AI-enabled self-learning models—and clarifies middleware requirements and representative deployment patterns. Second, it identifies critical enabling technologies and implementation pathways for trustworthy intelligent industrial ecosystems. Third, it highlights key future research directions, including model interpretability, edge-cloud collaboration, and cross-domain interoperability. The framework bridges theoretical advancement with practical industrial applicability, offering a scalable foundation for next-generation smart maintenance systems.

Existing digital twin (DT) research lacks a cross-level unified architecture and empirical validation across manufacturing tiers. Method: This paper proposes a generic DT architecture encompassing data flow, core components, and interaction mechanisms; classifies multi-tier DT types and modeling techniques; and critically analyzes discrete-event simulation (DES) for dynamic system modeling. Contribution/Results: Through industrial case studies, the work empirically validates DT’s efficacy in enhancing operational efficiency, reducing downtime, and optimizing lifecycle management. It identifies three critical implementation barriers: data integration complexity, cybersecurity risks, and high deployment costs. The study provides both theoretical foundations and practical guidelines for standardized DT development and scalable industrial deployment across product/production-line, production-system, and enterprise levels.

Integrating and enabling interoperability among heterogeneous digital models (DMs) from multiple sources in digital twins (DTs) faces persistent challenges—including the absence of standardized interfaces, high manual adaptation costs, and difficulties in cross-lifecycle model reuse—yet empirical, industry-grounded studies addressing these issues remain scarce. Method: This study conducts the first large-scale, cross-sectoral survey involving domain experts across diverse industries, complemented by expert interviews, thematic coding, and systematic requirement elicitation. Contribution/Results: We identify three fundamental bottlenecks hindering DM integration in DTs and propose semantic-driven interoperability and automated model orchestration as critical technical pathways forward. The findings constitute the first empirically grounded, industry-consensus-based evidence and technology roadmap for standardizing, automating, and semantically enabling DM integration within DT ecosystems.

To address the challenges of poor cross-instance sharing, low reusability, and inefficient collaborative configuration of digital twin (DT) assets, this paper proposes the first federated DT platform architecture designed for multi-user, multi-instance collaboration. Departing from traditional monolithic, siloed management paradigms, the architecture adopts a DT-as-a-Service (DTaaS) model and integrates service registration/discovery, semantic asset description, distributed configuration management, and a lightweight API gateway. This enables unified discovery, on-demand reuse, dynamic reconfiguration, and collaborative evolution of DT assets across heterogeneous domains. Evaluated in manufacturing and robotics scenarios, the platform achieves a 62% increase in asset reuse rate and reduces cross-instance collaborative configuration time by 57%, significantly enhancing development efficiency and system scalability.

This study addresses the deep integration of modeling and simulation (M&S) with artificial intelligence (AI) within digital twins to enhance their intelligent prediction and autonomous decision-making capabilities. By developing a comprehensive framework that incorporates physical modeling, discrete-event, and hybrid simulation methods alongside AI-driven advanced analytics and predictive modeling, the work elucidates the bidirectional synergy between M&S and AI. It positions the digital twin as a pivotal enabling platform for this convergence, revealing its multifaceted roles across business, development, and operational contexts. The research establishes an integrated theoretical foundation, identifies critical challenges, and outlines future directions to advance more intelligent and cohesive digital twin systems.

This study addresses the challenges of non-standardized, time-consuming, and resource-intensive digital twin development by proposing a tool-supported framework that, for the first time, enables the automatic derivation of purpose-specific digital twins from existing engineering models. By reusing pre-existing structural and behavioral models of physical assets and integrating lightweight customization and configuration, the approach leverages model-based engineering methods and automated generation techniques to rapidly align digital twins with operational objectives. The framework’s feasibility, generality, and engineering practicality are demonstrated through the successful automatic generation of digital twin instances across four heterogeneous use cases.

Digital twins are increasingly critical for real-time monitoring and decision-making in complex systems; however, their dynamic behavior, integration of multi-source heterogeneous data, and requirements for real-time synchronization pose significant challenges to verifying accuracy, reliability, and trustworthiness. To address these challenges, this paper proposes the first comprehensive, lifecycle-oriented TEVV (Testing, Evaluation, Verification, and Validation) framework for digital twins. The framework systematically integrates model-driven engineering, formal verification, simulation-based comparison, data provenance tracking, and uncertainty quantification, augmented with real-time monitoring and feedback mechanisms to enable multi-level, multi-dimensional trust assessment. Designed for cross-domain scalability, the framework has been empirically validated across multiple industrial applications, demonstrating substantial improvements in digital twin model credibility and decision-support effectiveness.

To address the loss of fidelity in digital twins caused by dynamic physical system evolution—such as maintenance, wear, and human intervention—this paper proposes a model-verification-based continuous validation framework. The framework integrates real-time monitoring with historical data comparison to construct an interpretable validation metric system, incorporates a lightweight anomaly detection mechanism, and introduces a data-driven parameter self-adaptation estimation algorithm for online twin diagnosis and closed-loop model updating. Unlike conventional static calibration methods, our approach enables long-term trustworthiness preservation and autonomous evolution of the digital twin. Evaluated on an industrial quay crane use case, the framework accurately detects system deviations and dynamically refines model parameters, reducing modeling error by 37.2% and improving maintenance response timeliness by 52%. These results demonstrate significant enhancements in the representativeness, robustness, and engineering practicality of digital twins.

Current digital twin (DT) and simulation platform integrations in IoT and IIoT suffer from rigidity and insufficient runtime coordination, hindering adaptive system operation and closed-loop interaction between virtual models and physical assets. Method: This paper proposes a bidirectional integration framework centered on a novel “Digital Twin–Simulation Bridge” mechanism, enabling dynamic model updates, parameter self-adaptation, and deep integration of virtual commissioning with real-time behavioral analysis. The framework adopts a modular architecture and standardized bidirectional data interfaces to support flexible, scalable interconnection across the DT lifecycle—encompassing design, verification, and real-time execution—with heterogeneous simulation platforms. Contribution/Results: Experimental evaluation demonstrates that the framework significantly enhances design agility and enables high-fidelity, closed-loop co-simulation and physical-device synchronization across diverse industrial scenarios, thereby improving operational responsiveness and system adaptability.

Efficient, secure, and interoperable synchronization between digital twins and their physical counterparts remains a significant challenge. This work proposes a standardized, data-centric synchronization architecture tailored for industrial applications, which, for the first time, systematically integrates existing synchronization techniques, middleware, and low-latency communication mechanisms into a unified framework. By harmonizing these components, the architecture not only ensures robust security and cross-platform interoperability but also substantially enhances synchronization accuracy and resilience. The proposed approach offers a scalable and practical pathway toward realizing dependable digital twin systems, effectively addressing a critical gap in current technological capabilities.

Digital twins in healthcare face critical bottlenecks including poor interoperability, high data privacy risks, and insufficient model fidelity. To address these challenges, this study proposes a full-lifecycle medical digital twin framework integrating multimodal real-time data (e.g., medical imaging, biosensor streams), explainable AI, and federated learning to construct a high-fidelity, multi-organ协同 virtual physiological model. Crucially, it is the first to incorporate genomic data and an embedded ethical governance mechanism. The framework enables three novel applications: (1) cardiac functional simulation, (2) tumor radiotherapy response prediction, and (3) pharmacokinetic modeling. Validation demonstrates significant improvements in prognostic accuracy, drug discovery efficiency, and clinical decision support. By unifying heterogeneous data sources, preserving privacy via decentralized learning, and embedding biological interpretability and ethical oversight, this work advances medical digital twins beyond single-disease modeling toward a predictive, preventive, and personalized healthcare paradigm.

This work addresses the interoperability and runtime coordination challenges among heterogeneous digital twins in federated ecosystems, which arise from divergent modeling approaches and technology stacks. To overcome these issues, the paper proposes a federated node manager architecture that enables modular and deployable integration of digital twins through capability-controlled exposure, protocol and data schema adaptation, and real-time exchange of states and events. Built upon service-oriented and event-driven design principles, the proposed system demonstrates the feasibility of efficient collaboration among diverse digital twins in an intelligent transportation emergency response scenario. This implementation successfully translates conceptual federated architectures into a practical, operational technical pathway, marking the first realization of such an integrated solution in real-world settings.