IoT and AI Models

This chapter bridges AI theory with real-world deployment of intelligence at the industrial edge. It introduces Industry 5.0, a human-centric paradigm uniting sustainable automation with collaborative intelligence and maps the enterprise IoT stack - from connectivity layers and real-time streaming pipelines to analytics-ready data lakes. With this foundation, two end-to-end machine learning workflows are constructed to detect subtle single- and three-phase faults in AC current-voltage waveforms, demonstrating stepwise signal processing, interquartile-range (IQR) heuristics, and resilient pattern recognition amid noisy conditions. The discussion then shifts to Small Language Models (SLMs), illustrating how lightweight embeddings and Hierarchical Navigable Small-World (HNSW) indexing enable efficient ESG-aware search on low-power edge gateways. Practical Docker recipes and C++ bindings conclude the chapter, showing how to containerize, deploy, and query these models securely and at scale. By integrating IoT telemetry, edge analytics, and human oversight, readers gain both conceptual clarity and hands-on techniques for realizing the promise of Industry 5.0–where intelligence, sustainability, and resilience converge at the edge