Research on the Efficiency Improvement Model of Lighthouse Factory Operation Based on XGBoost Algorithm

This study proposes an operational efficiency improvement model for Lighthouse Factories based on the XGBoost algorithm. By integrating large-scale heterogeneous data from PLCs, MES, SCADA systems, maintenance records, and energy monitoring platforms, the model identifies operational inefficiencies and predicts critical production events through advanced feature engineering and cost-sensitive learning strategies. A comprehensive data-to-decision framework is established, enabling predictive maintenance, dynamic scheduling, and near real-time decision support. Experimental evaluation on an 18-month industrial dataset containing over 15 million records demonstrates superior performance compared with conventional machine learning approaches, achieving higher predictive accuracy, reduced unplanned downtime, and improved production throughput. The proposed framework is particularly applicable to Industry 4.0 environments supported by industrial wireless communication networks, antenna-enabled Industrial Internet of Things (IIoT) infrastructures, and cyber-physical manufacturing systems, where reliable data acquisition and transmission are essential for intelligent operation. The results confirm the effectiveness, scalability, and practical value of the proposed model for enhancing operational efficiency in smart manufacturing environments.