A Multi-Objective Optimization Framework for Energy–Efficiency Balance in Intelligent Tobacco Logistics Parks Based on NSGA-II

The increasing demand for sustainable logistics management in the tobacco industry raises a key scientific question: how to model and optimize the dynamic balance between energy consumption and operational efficiency within large-scale, cyber–physical logistics systems. To address this challenge, this paper proposes a multi-objective optimization framework based on the Non-dominated Sorting Genetic Algorithm II (NSGA-II), integrated with digital twin and IoT technologies. The framework constructs a real-time virtual–physical synchronization mechanism, enabling adaptive decision-making under fluctuating workloads and uncertain energy conditions. By formulating energy–efficiency coordination as a bi-objective optimization problem, the research advances the scientific understanding of multi-objective evolutionary dynamics in industrial logistics contexts. Taking the Zhaoqing Tobacco Intelligent Logistics Park in Guangdong Province, China, as a case study, the proposed model integrates IoT sensing data, energy monitoring, and logistics flow optimization under real-time constraints. The optimization framework formulates a bi-objective problem, minimizing total energy consumption while maximizing logistics throughput. By employing digital twin modeling, the system continuously synchronizes virtual and physical park states, enabling dynamic scheduling and predictive control. The NSGA-II algorithm is enhanced through adaptive crowding distance adjustment and variable mutation rates to handle time-varying logistics workloads. Experimental results using actual operational data from the Zhaoqing park demonstrate that the proposed method achieves up to 17.8% (from 3 600 kWh to 2 960 kWh per day) energy reduction and 12.3% improvement in throughput compared with traditional rule-based and standard GA methods compared to rule-based baseline. The findings indicate that NSGA-II provides an effective and scalable approach for real-time decision-making in energy-aware logistics operations. This research contributes to the development of intelligent, low-carbon logistics systems and aligns with China’s “Dual Carbon” (carbon peak and neutrality) strategic objectives.