Blockchain-Enabled Secure Financial Management Model with IoT-Based Real-Time Asset Monitoring

Abstract

Recent advances in the Internet of Things (IoT), artificial intelligence (AI), and blockchain have enhanced enterprise financial monitoring through real-time analytics, predictive decision-making, and secure data management. Previous research mainly relies on single anomaly detection models, limiting the capture of temporal financial behaviour, statistical irregularities, and secure risk verification in complex industrial environments. The integration of the LSTM Autoencoder, Isolation Forest, and Hyperledger Fabric enables accurate financial anomaly detection, secure blockchain-based data management, and reliable enterprise decision-making through continuous IoT-based asset monitoring in dynamic operational environments. The objective of this study is to develop an intelligent Financial-IoT framework for accurate financial risk detection and secure blockchain-based decision support. The proposed methodology uses a Financial–IoT dataset comprising 100,000 records of enterprise financial indicators and IoT operational metrics. Data preprocessing consists of temporal splitting and Principal Component Analysis (PCA), which keeps 95% of the variance with 9 principal components. It combines the LSTM Autoencoder for temporal anomaly detection, Isolation Forest for statistical outlier detection, Hyperledger Fabric, and AES-256 and SHA-256 for secure decentralised validation. The experimental results show that it outperforms with 97% classification accuracy, 98% precision, 97% recall, 97% F1 score and 99% ROC-AUC. The system ensures a prediction latency of less than 50ms and throughput of over 1800 transactions per second. The results show that the proposed hybrid approach enhances the accuracy of anomaly detection, scalability, transparency, and security of financial management applications in the intelligent enterprise.