Abstract
A comparison of the use of big data computing in Industry 4.0 and Industry 5.0 was carried out utilizing real-world data to evaluate production effectiveness, quality control, and energy consumption. The findings suggest that Industry 5.0 paradigms result in a 2% drop in the number of faulty items produced and a 1% decrease in energy usage in highly automated companies. Quality control analysis revealed that "Type B" faults account for approximately 65% of total defects in Industry 4.0 settings. These results highlight the benefits of Industry 5.0, which capitalizes on human-machine cooperation and data-driven processes to create more efficient, sustainable, and quality-driven manufacturing environments.
Introduction
While the transition from Industry 4.0 (focused on automation and IoT) to Industry 5.0 (focused on human-machine collaboration and personalization) is a significant trend in manufacturing, empirical data comparing the two is scarce. Most existing literature focuses on theoretical frameworks rather than real-world experimental results. This study aims to bridge this gap by investigating how big data computing functions across both paradigms, specifically looking at key performance indicators (KPIs) like production effectiveness, quality control, and energy consumption to determine if the human-centric approach of Industry 5.0 provides tangible benefits in a factory setting.
Methodology
The researchers employed a structured experimental methodology to gather and analyze data from industrial environments. The approach included simultaneous data collection from production facilities representing both Industry 4.0 and 5.0 contexts (labeled F1 and F2) between January and May 2023.
The datasets included production metrics (total products, defects, costs), operational data (automation levels, energy consumption), quality control logs (defect types), and IoT sensor data (temperature, humidity, vibration). Statistical analysis, including t-tests, ANOVA, and regression analysis, was used to determine the significance of the differences between the two industrial stages.
Results and Discussion
Quality Control and Defect Reduction
A key finding of the study is the efficiency gain provided by human-machine cooperation. Contrary to the assumption that human involvement might slow processes, the collaborative nature of Industry 5.0 actually improved quality. Industry 5.0 saw a 2% decrease in the overall number of faulty items compared to Industry 4.0. While Industry 4.0 (Factory F1) averaged higher defect rates, Industry 5.0 (Factory F2) dropped the average to 190 faulty items per day.
Furthermore, the study identified that "Type B" defects accounted for roughly 65% of all issues in Industry 4.0 settings. This highlights a bottleneck in traditional automated quality control that Industry 5.0's more adaptive techniques—such as augmented reality and collaborative robots—are better equipped to handle.
Energy Efficiency and Environmental Stability
Highly automated companies saw a 1% decrease in energy usage when moving to Industry 5.0 frameworks, averaging 4,950 kWh/day compared to 5,000 kWh/day in Industry 4.0. This improvement is linked to superior environmental regulation. The Industry 5.0 environment showed highly stable IoT sensor readings, with temperature and humidity remaining within a narrow 2% range.
Industry 5.0 sensors maintained an average temperature of 21.6°C and humidity of 46.4%, with vibration levels averaging 1.34 mm/s². This stability is a direct contributor to the reduction in faulty items, suggesting that the enhanced data ecosystems of Industry 5.0 provide superior environmental control compared to rigid Industry 4.0 automation.
Conclusion
The shift toward Industry 5.0 paradigms leads to measurable gains in quality and efficiency. By leveraging human-machine synergy and advanced data-driven processes, manufacturers can achieve more sustainable and quality-driven outcomes. Future research should explore the specific roles of collaborative robots (cobots) and how different levels of human participation affect the speed of the "data-to-decision" cycle across various manufacturing scales.
Summary updated