期刊論文
| 學年 | 114 |
|---|---|
| 學期 | 2 |
| 出版（發表）日期 | 2026-06-03 |
| 作品名稱 | Stabilizing Distributed Financial Control Loops In Industry 4.0 Via Deterministic Wireless Digital Twins |
| 作品名稱（其他語言） | |
| 著者 | Yin-Yin Huang; Jingchao Pan; Hsien-Yu Chen; Chien-Ming Huang |
| 單位 | |
| 出版者 | |
| 著錄名稱、卷期、頁數 | IEEE Communications Standards Magazine ,p.1-9 |
| 摘要 | Industry 4.0 environments are increasingly characterized by interconnected cyber-physical systems, autonomous machines, and financial transactions. In such systems, distributed financial control loops governing pricing, billing, cash allocation, and supply-chain payments are prone to instability due to network delays, asynchronous updates, and unpredictable communication jitter. This research proposes a Deterministic Wireless Digital Twin framework (DWDTF) to stabilize financial control loops by integrating digital twins, deterministic wireless networking, and AI-driven decision support. The framework creates multi-layered digital twins representing physical assets, production processes, and financial variables, continuously synchronized over bounded-latency, ultra-reliable wireless networks. In practical enterprise environments, financial systems often operate using periodic batch reconciliation processes, and the proposed framework can function in a hybrid mode that integrates real-time monitoring with scheduled reconciliation cycles. A dataset with 1,000 records and 14 features capturing financial, operational, and network parameters is employed. Data preprocessing includes cleaning, and Min-Max normalization to scale variables for stable and efficient learning. See-See Partridge Chicks Optimized Least Squares Support Vector Machine (SSPC-LS-SVM) is employed to predict financial fluctuations, detect anomalies, and recommend adaptive control actions, while deterministic networking ensures timely execution of corrective decisions. In practice, certain high-impact financial decisions may incorporate validation checkpoints or controlled delays to ensure compliance with risk management and organizational policies. SSPC ensures stable and optimal tuning of control parameters, while LS-SVM accurately predicts financial fluctuations and detects anomalies for timely corrective actions. Results demonstrate high predictive and control performance, with accuracy, recall, and F1-scores ranging from 94.67% to 95.86%, outperforming existing methods. By integrating predictive AI, digital twins, and deterministic time-sensitive communication, the system enables proactive, resilient financial management, aligning economic decisions with production processes, although financial objectives and operational goals may at times conflict, requiring trade-off analysis and coordinated decision strategies within the framework and providing a scalable, flexible, data-driven architecture for stable, autonomous Industry 4.0 operations. |
| 關鍵字 | |
| 語言 | en_US |
| ISSN | |
| 期刊性質 | 國外 |
| 收錄於 | ESCI Scopus ABS* |
| 產學合作 | |
| 通訊作者 | |
| 審稿制度 | 否 |
| 國別 | TWN |
| 公開徵稿 | |
| 出版型式 | ,電子版 |
| 相關連結 |
機構典藏連結 ( http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/129644 ) |