International Journal of Advanced Research in Computer and Communication Engineering

Abstract: Manufacturing industries are embracing smart solutions to achieve operational excellence by enhancing controllability, visibility, and flexibility. An intelligent quality control system with autonomous capabilities is a critical enabler of smart manufacturing. The autonomous quality control mechanism integrates edge artificial intelligence and cloud orchestration—AI-based applications for real-time anomaly detection and predictive analysis, AI-enabled cloud services for system orchestration, and edge-cloud data governance. The effectiveness of the proposed approach is demonstrated in a case study involving a complex process using Internet-of-Things devices for data acquisition.
Understanding the complex, repetitive, and noisy nature of manufacturing processes with advanced machine-learning algorithms often requires substantial data-analytics infrastructure. Moving all data to the cloud for processing and storing is not feasible for operational efficiency when the core functions are repetitive, time-sensitive, and performance-critical. Solutions deployed on edge devices provide limited performance and efficiency due to challenged computing resources. An edge–cloud quality-control framework with real-time anomaly detection and predictive analytics capabilities is proposed. AI-based applications with active learning on the edge provide constant real-time services to detect data anomalies in quality features from quality-control check points. AI-enabled cloud services orchestrate the entire system by continuously monitoring operational conditions and storing all data, validating the use of predictive-quality-control analysis.

Keywords: Smart Manufacturing Systems, Autonomous Quality Control, Intelligent Quality Inspection, Edge Artificial Intelligence, Cloud-Based AI Orchestration, Edge–Cloud Integration, Real-Time Anomaly Detection, Predictive Quality Analytics, Industrial Internet of Things (IIoT), AI-Driven Process Monitoring, Active Learning at the Edge, Manufacturing Data Governance, Distributed AI Architectures, Time-Critical Industrial Analytics, Operational Excellence Enablement, Scalable Quality-Control Frameworks, Edge–Cloud Data Pipelines, Performance-Critical AI Systems, Intelligent Manufacturing Operations, AI-Orchestrated Industrial Systems.

Downloads: | DOI: 10.17148/IJARCCE.2024.131271