International Journal of Advanced Research in Computer and Communication Engineering

Abstract: In modern manufacturing environments, ensuring consistent product quality is critical for maintaining competitiveness and customer satisfaction. This study proposes an online machine learning-based approach for real-time industrial product quality analysis. Unlike traditional offline models that require retraining on static datasets, the online learning paradigm enables the system to continuously update itself with incoming data, adapting to process variations and new patterns without significant downtime. The proposed system integrates streaming data from production lines—such as sensor readings, visual inspections, and process parameters—to predict and detect quality anomalies early in the production cycle. Experimental results on industrial datasets demonstrate that the online learning models, including algorithms like Online Gradient Descent and Adaptive Random Forest, achieve high accuracy and robustness while reducing latency in decision-making. This approach enhances operational efficiency, minimizes defective output, and supports predictive maintenance strategies in Industry 4.0 settings.With the growing adoption of Industry 4.0 technologies, real-time quality monitoring has become essential in modern manufacturing systems. This study presents an online machine learning- based approach for industrial product quality analysis, aimed at improving defect detection and process optimization. Unlike traditional batch learning models, online machine learning algorithms continuously update their parameters with streaming data, allowing for adaptive learning in dynamic production environments. The system leverages real-time data from sensors and inspection tools to identify deviations in product quality as they occur.
Algorithms such as Online Gradient Descent and Adaptive Random Forest were evaluated for their performance in handling non-stationary data streams. Experimental results show that the proposed method provides high accuracy, low latency, and efficient resource usage, making it suitable for deployment in smart manufacturing systems. This research highlights the potential of online learning to enhance product reliability, reduce waste, and support intelligent decision- making in industrial processes.

Keywords: Online Machine Learning Industrial Quality Control, Real-time Monitoring, Smart Manufacturing, Adaptive Algorithms, Product Defect Detection, Industry 4.0, Data Streams, Predictive Maintenance, Machine Learning in Manufacturing.

Downloads: | DOI: 10.17148/IJARCCE.2025.1411104