Optimization of short-haul cargo volume prediction and scheduling based on XGBoost and robust optimization
DOI:
https://doi.org/10.54097/cq5cnv59Keywords:
Short-haul Transportation, Cargo Volume Prediction, XGBoost, Robust Optimization, Scheduling Optimization.Abstract
Short-distance transportation plays an important role in modern logistics network, especially in the end distribution, which directly affects the customer experience and logistics efficiency. This paper focuses on the issues of cargo volume prediction and vehicle scheduling in short-distance transportation. It proposes corresponding solutions to three main problems and optimizes transportation efficiency and cost by integrating XGBoost and robust optimization techniques. First, for the cargo volume prediction problem, this paper constructs a prediction model based on time series features using XGBoost algorithm to predict the cargo volume of each route in the next 24 hours, and further refines the results to a 10-minute granularity, which provides a fine-grained data support for the subsequent scheduling optimization. Secondly, based on the prediction results, this paper establishes a vehicle scheduling model, which determines the transportation demand and shipping time under the premise of shipping node constraints; meanwhile, through the preferential use of owned vehicles and the design of string-point scheme, it realizes the improvement of transportation efficiency and total cost. Finally, to address the uncertainty caused by prediction deviation, this paper introduces robust optimization and scenario analysis methods to evaluate the scheduling effect under different deviations, ensuring that the stability and low cost of the system can be maintained in the most unfavorable scenarios. By combining XGBoost and robust optimization methods, this study proposes a comprehensive solution that can effectively improve the efficiency of the short-haul transportation system, external carrier dependence, and ensure the robust operation of the scheduling system under the deviation of the cargo volume forecast. This research not only has important theoretical value, but also has strong practical application prospects.
Downloads
References
[1] Chen Zhi, Fan Wei. A Travel Time Prediction Method for Expressways Based on XGBoost Model [J]. Sustainability, 2021, 13 (15): 8577.
[2] Gutmann S, Makridis M, Ciuffo B. Truck parking occupancy prediction with XGBoost-LSTM model [J]. Future Transportation, 2021, 1 (1): 23 - 40.
[3] Guo Dan. Multi-objective Optimization Research on Short-distance Transportation Scheduling Based on NSGWO [J]. Sustainability, 2024, 16 (15): 6310 - 6325.
[4] Wang Fang, Li Qiang, Zhou Jie. Urban Distribution Demand Forecasting Based on Improved LSTM [J]. Computer Applications, 2020, 40 (8): 2285 - 2291.
[5] Wang Kai, Liu Tong. Research on Prediction of Parcel Volume in Last-mile Logistics Based on Deep Learning [J]. Logistics Technology, 2021, 40 (6): 112 - 116.
[6] Li Ming, Zhao Lei, Zhang Jing. Research on Optimization of Short-distance Transportation Vehicle Scheduling for E-commerce Logistics [J]. Systems Engineering - Theory & Practice, 2019, 39 (12): 3031 - 3043.
[7] Polinder G J, Veelenturf L P, Kroon L G. An adjustable robust optimization approach for periodic railway timetabling [J]. Transportation Research Part B, 2019, 123: 249 - 268.
[8] Liu Yang, Luo Xiang, Wei Xiang, et al. Research on Robust Optimization Model for Dynamic Scheduling of Single-Line Buses [J]. Sustainability, 2022, 14 (1): 73 - 85.
[9] Ghanbari S R, Afshar-Nadjafi B, Sabzehparvar M. Robust optimization of train scheduling with consideration of response actions [J]. Mathematical Biosciences and Engineering, 2023, 20 (7): 13015 - 13035.
[10] Wang Lei, Zhang Hong. Research on the Application of Robust Optimization in Urban Public Transport Scheduling [J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20 (2): 12 - 20.
[11] Shehadeh K S. Distributionally robust optimization approaches for a stochastic mobile facility fleet sizing, routing, and scheduling problem[J]. Transportation Science, 2023, 57 (2): 489 - 509.
[12] Agra A, Andersson H, Christiansen M, et al. Robust optimization for a maritime inventory routing problem [J]. Transportation Science, 2018, 52 (3): 509 - 525.
[13] Lin Feng. Review of Distributed Robust Optimization Methods: Theory and Applications [J]. Mathematics in Practice and Theory, 2022, 52 (14): 45 - 59.
[14] Zhang Y, He S, Wang J. Gradient boosting decision tree for freeway travel time prediction [J]. IEEE Access, 2020, 8: 143390 - 143399.
[15] Li B, Bai Y. Gradient boosted freight travel time prediction in Ningbo [J]. Journal of Advanced Transportation, 2019, 2019: 1 - 10.
[16] Gupta R, Kumar A, Dey L. Taxi travel time prediction using random forest and gradient boosting [J]. Procedia Computer Science, 2019, 152: 735 - 742.
[17] Kunichetty N. Design of transportation volume forecasting model [J]. International Journal of Forecasting, 2021, 37 (4): 1581 - 1593.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Highlights in Science, Engineering and Technology
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
