Citation :

Shankar Narayanan SGS, 2025. "Hybrid Edge AI and Centralized Processing for IoT: A Scalable, Secure Framework for Real-Time Manufacturing Analytics" ESP International Journal of Advancements in Computational Technology (ESP-IJACT)  Volume 2, Issue 2: 126-131.

Abstract :

The convergence of Industry 4.0 and IoT has fueled a need for low-latency decision-making at the edge of production systems, while also leveraging centralized resources for global analytics and security. This paper proposes a Hybrid Edge-Central (HEC) Architecture that partitions Graph Convolutional Networks (GCNs) across on-premises devices and cloud data centers, supplemented by Generative Adversarial Networks (GANs) for adversarial testing. A hardware security module (HSM) on each edge device protects API keys and ensures tamper-resistant cryptographic operations. We demonstrate how edge AI reduces latency and bandwidth usage (~60%), while cloud-side GCN layers capture cross-device correlations for predictive maintenance and anomaly detection. A case study in an automotive manufacturing plant showcases significant reductions in unplanned downtime (~30%) and improved security via GAN-driven stress testing. Our results highlight a robust, scalable solution for Industry 4.0 transformations requiring data security, real-time insights, and system-wide optimization.

References :

[1] ISO 27001 (2013). Information technology — Security techniques — Information security management systems — Requirements. International Organization for Standardization.

[2] Lane, N. D., Bhattacharya, S., Georgiev, P., et al. (2019). DeepX: A Software Accelerator for Low-Power Deep Learning Inference on Mobile Devices. In Proceedings of MobiSys.

[3] NIST (2018). Framework for Improving Critical Infrastructure Cybersecurity. National Institute of Standards and Technology.

[4] Shi, W., Cao, J., Zhang, Q., Li, Y., & Xu, L. (2020). Edge Computing: Vision and Challenges. IEEE Internet of Things Journal, 3(5), 637–646.

[5] Wu, J., Lee, J., & Zhao, W. (2020). Fault Detection and Diagnosis in Smart Manufacturing Using GCN. IEEE Transactions on Industrial Informatics, 16(6), 3705–3716.

[6] Zhang, J., Zhang, H., & Tao, C. (2021). Adversarial Testing of Autonomous Driving via GAN-Generated Scenarios. IEEE Transactions on Intelligent Transportation Systems, 22(7), 4234–4243.

Keywords :

IoT, Artificial Intelligence, Framework, Manufacturing Analytics.