Academic Journal of Computing & Information Science, 2025, 8(11); doi: 10.25236/AJCIS.2025.081110.
Yating Zhong1, Jianchao Yu2, Jinkun Chen2, Guorui Zhao3
1School of Business, Guangdong Ocean University, Yangjiang, China
2School of Mechanical and Energy Engineering, Guangdong Ocean University, Yangjiang, China
3School of Computer Science and Engineering, Guangdong Ocean University, Yangjiang, China
Urban waste classification transportation route optimization, as a typical vehicle routing problem (VRP), is of critical significance for improving urban management efficiency. This paper constructs a progressive optimization model based on graph theory and mixed integer programming, and solves it by combining a weighted graph approximation algorithm, an improved simulated annealing algorithm, and multi-objective adaptive large neighborhood search (MOALNS) algorithm. The system addresses three core problems: single-type garbage path planning, multi-type garbage co-transportation, and a multi-objective optimization with transfer stations. The case verification results show that the single-objective basic model can achieve the optimal driving distance of 1,140.00 km for 16 vehicles, the multi-vehicle collaborative model can control the daily transportation cost at 3,123.02 yuan, and the multi-objective joint model can reduce the total cost to 2,460.15 yuan and carbon emissions to 945.80kg through the site selection optimization of three transfer stations. The comparative experiments show that the MOALNS algorithm is significantly superior to traditional algorithms such as NSGA-II and MOEA/D in terms of cost convergence. Taking the P-n23-k8 dataset as an example, its HV value reaches 0.771, and the multi-objective optimization accuracy advantage is prominent. Research shows that graph-theoretic methods and improved heuristic algorithms can effectively enhance the economic efficiency and environmental friendliness of garbage transportation, among which the MOALNS algorithm provides efficient support for multi-objective optimization in complex scenarios.
Garbage classification and transportation, Path optimization, Vehicle routing problem, MOALNS algorithm, Multi-objective optimization
Yating Zhong, Jianchao Yu, Jinkun Chen, Guorui Zhao. Research on Optimization and Scheduling Models and Algorithms for Urban Waste Sorting Transport Routes. Academic Journal of Computing & Information Science (2025), Vol. 8, Issue 11: 89-98. https://doi.org/10.25236/AJCIS.2025.081110.
[1] Li W, Wang P, Xu Y, et al. Multi-Objective Optimization of Municipal Solid Waste Collection Based on Adaptive Large Neighborhood Search[J]. Electronics,2024,14(1): 103.
[2] Lei Qin, Gao Yanbing, Zhou Yufeng, et al. Multi-Delivery Option Route Planning Based on an Improved ALNS Algorithm [J]. Systems Engineering and Electronics, 2025, 47(1): 173–181.
[3] Santini A, Ropke S, Hvattum L M. A comparison of acceptance criteria for the adaptive large neighbourhood search metaheuristic[J]. Journal of Heuristics,2018,24(5): 783–815.
[4] Yang Xiuwen, Chen Zhenjie, Li Ailing, et al. Determining Optimal H-Circles Using Matrix Flip Method [J]. Journal of Logistics Engineering University, 2008, 24(1): 102–106.
[5] Li Lixin, Tan Jianrong. A multi-diagonal exchange algorithm for forcibly embedding constraint edges in constrained Delaunay triangulation[J]. Acta Informatica Sinica, 1999, 22(10): 1114-1118.
[6] Chen Huagen, Wu Jiansheng, Wang Jialin, et al. Research on the Mechanism of Simulated Annealing Algorithms [J]. Journal of Tongji University (Natural Science Edition), 2004, 32(6): 802-805.