Cloud–Edge Integrated Digital Twin Architecture for Predictive Analytics in Smart Infrastructure

DOI

10.2991/978-94-6239-693-7_37How to use a DOI?

Keywords

Digital Twin; Cloud–Edge Computing; Predictive Analytics; Smart Infrastructure; Cyber-Physical Systems; Edge Intelligence

Abstract

Transportation, energy, water, and built environments are increasingly being used as more innovative infrastructure through predictive analytics to guarantee reliability, efficiency, and resilience. Conventional centralized data analytics platforms have limitations on scalability, latency, and privacy when used on real-time infrastructure monitoring and forecasting. In this paper, a proposal is made concerning a cloud-edge integrated digital twin architecture to support distributed data processing, real-time synchronization, and predictive intelligence on the smart infrastructure systems. The architecture integrates edge-level data ingestion and preprocessing with cloud-level model training, simulation, and optimization to continuously align physical assets and their digital replicas. The suggested framework is assessed using a multi-domain infrastructure dataset containing 2.4 million time-stamped sensor messages from transportation, energy, and environmental monitoring systems. Five metrics are used to measure predictive performance: root mean square error, mean absolute percentage error, prediction latency, system throughput, and model update delay. It was experimentally demonstrated that the proposed architecture can achieve 37% lower prediction latency and 22% higher forecasting accuracy than cloud-only digital twin systems. The edge-cloud synchronization scheme allows almost real-time model updates with an average propagation time of 1.8 s, allowing quick adjustment to the infrastructure dynamics. The statistical analysis shows that decentralized preprocessing reduces network load by 41% without compromising model fidelity (correlation above 0.94 with centralized baselines). A study involving an ablation shows that deleting edge analytics increases latency by 29%, whereas deleting cloud-level optimization decreases long-term prediction accuracy by 17%. The findings suggest that cloud-edge-built digital twins offer a scalable and robust platform for predictive analytics in intelligent infrastructure, enabling proactive maintenance, risk reduction, and sustainable urban management.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Anjali Krushna Kadao
AU  - Moti Ranjan Tandi
PY  - 2026
DA  - 2026/06/16
TI  - Cloud–Edge Integrated Digital Twin Architecture for Predictive Analytics in Smart Infrastructure
BT  - Proceedings of the International Conference on Intelligent Systems for a Sustainable Future (ISSF 2026)
PB  - Atlantis Press
SP  - 365
EP  - 374
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6239-693-7_37
DO  - 10.2991/978-94-6239-693-7_37
ID  - Kadao2026
ER  -