# Seamless Deployment of Machine Learning Inference at the Far Edge: An Empirical AWS IoT Greengrass Architecture for Industrial Predictive Maintenance

arXiv: https://arxiv.org/abs/2610.14815 · HTML: https://mlsanity.com/paper/2610.14815/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 30 references that are in the map.
- First posted on arXiv 10 Oct 2026.

## Builds on (most cited first, 5 of 5)

- [Applications of Distributed Machine Learning for the Internet-of-Things: A Comprehensive Survey](https://mlsanity.com/paper/2310.10549/) (2023)
- [ECAvg: An Edge-Cloud Collaborative Learning Approach using Averaged Weights](https://mlsanity.com/paper/2310.03823/) (2023)
- [JANUS: Benchmarking Commercial and Open-Source Cloud and Edge Platforms for Object and Anomaly Detection Workloads](https://mlsanity.com/paper/2012.04880/) (2020)
- [Enhancing failure prediction in nuclear industry: Hybridization of knowledge- and data-driven techniques](https://mlsanity.com/paper/2511.11604/) (2025)
- [Self-Evolving Multi-Agent Network for Industrial IoT Predictive Maintenance](https://mlsanity.com/paper/2602.16738/) (2026)

## Abstract (arXiv)

Industrial IoT sensors produce continuous, high-volume data streams that predictive maintenance needs to analyse in real time. Cloud-centric architectures send all raw telemetry to central servers, which costs latency, bandwidth and resilience. This paper presents a hybrid cloud-edge architecture for manufacturing predictive maintenance built on AWS IoT Core and AWS IoT Greengrass v2, with local machine learning inference at the edge and selective cloud synchronization. We deploy two configurations on real AWS infrastructure and compare them: a cloud-only baseline (Configuration A) and the proposed hybrid edge architecture (Configuration B). We measure end-to-end inference latency, bandwidth consumption and inference accuracy. Configuration B cuts cloud-bound bandwidth by 86.8% (63.6 KB to 7.2 KB per 200-reading window) and decision latency by 256x (23.1 ms cloud round trip to 0.09 ms local inference). At the edge, with no cloud involvement, it keeps an anomaly-class F1 score of 0.84 and 91% recall. The evaluation also instruments system uptime and Stream Manager queuing under cloud disconnection; the quantitative disconnection experiment is left as planned work. Our review of the literature points to four gaps that this work addresses: the lack of empirically validated Greengrass v2 deployments for predictive maintenance, unquantified bandwidth savings from selective edge synchronization, unmeasured resilience under cloud disconnection, and the open MLOps feedback loop at the edge.