Time-Series Classification Using AI Models for Digital Twin Applications

Digital Twin (DT) technology enables real-time monitoring and optimization through virtual replicas that stay synchronized with physical systems. This paper reviews AI-based time-series classification methods for DT applications, covering both classical machine learning and deep learning architectures, including CNNs, RNNs, TCNs, and Transformers. We discuss integration challenges such as real-time requirements, edge-cloud deployment, and multimodal data fusion, and highlight applications across manufacturing, healthcare, and infrastructure. Key challenges, including data scarcity, interpretability, and scalability, are identified as important directions for future research.