快递揽件需求区域时空预测方法研究

Spatiotemporal Forecasting for Regional Express Pickup Demand

摘要

为了提高快递揽件环节效率，识别高需求区域成为亟待解决的问题。基于区域内兴趣点数据，使用层次图信息最大化 (hierarchical graph infomax, HGI) 方法得到区域嵌入，将其与历史揽件需求序列作为图多头注意力网络 (graph multi-head attention network, GMAN) 的输入，提出 HGI-GMAN 组合模型。在某快递公司真实数据集上的实验结果表明，与 5 种经典基准模型相比，HGI-GMAN 在回归指标 (RMSE、R²) 和分类指标 (macro-F1) 上均取得了较好效果。超参数敏感性分析和消融实验验证了所提出组合模型的稳健性和特征的有效性。

Abstract

To improve the efficiency of the parcel pickup operations, accurately identifying high-demand areas has become a key challenge. Based on point-of-interest data, the hierarchical graph infomax (HGI) method is employed to generate regional embeddings. These embeddings, along with historical pickup demand series, are then input into a graph multi-head attention network (GMAN) for prediction, thereby forming the proposed HGI-GMAN model. Experimental results on a real-world dataset from an express company indicate that the HGI-GMAN model outperforms five classical baseline models across various regression (RMSE, R²) and classification (macro-F1) metrics. Additionally, hyperparameter sensitivity analysis and ablation studies verify the robustness of the proposed model and the effectiveness of the extracted features.

关键词

快递行业 / 需求预测 / 时空序列预测 / HGI-GMAN 模型 / 深度学习

Key words

express industry / demand prediction / spatiotemporal forecasting / hierarchical graph infomax-graph multi-head attention network (HGI-GMAN) model / deep learning

参考文献

[1] 董成祥, 魏昕, 张坤鹏, 等 . 基于 Graph Transformer 的大规模路网交通流量预测[J]. 工业工程, 2023, 26(5): 159-167.

[2] Dong Chengxiang, Wei Xin, Zhang Kunpeng, et al. Traffic flow prediction for large-scale road network based on graph transformer[J]. Industrial Engineering Journal, 2023, 26(5): 159-167.

[3] 董成祥, 魏昕, 张坤鹏, 等 . 基于图卷积网络的乘客打车需求预测[J]. 工业工程, 2022, 25(5): 98-105.

[4] Dong Chengxiang, Wei Xin, Zhang Kunpeng, et al. Passenger ride-hailing demand prediction based on graph convolutional networks[J]. Industrial Engineering Journal, 2022, 25(5): 98-105.

[5] Wang S, Chen A, Wang P, et al. Predicting electric vehicle charging demand using a heterogeneous spatiotemporal graph convolutional network[J]. Transportation Research Part C: Emerging Technologies, 2023, 153: 104205.

[6] 李辰颖. 基于 CEEMD-SVM 组合模型的快递业务量预测[J]. 统计与决策, 2019, 35(12): 83-85.

[7] 许可, 王孝坤. 基于标准差法组合预测模型的区域快递需求量预测研究——以大连市为例[J]. 数学的实践与认识, 2022, 52(9): 18-27.

[8] Xu Ke, Wang Xiaokun. Regional express delivery based on standard deviation combined forecasting model demand forecasting research-taking Dalian city as an example[J]. Mathematics in Practice and Theory, 2022, 52(9): 18-27.

[9] 薛蓉娜, 张明敏, 南玉婷, 等 . 基于 GRU 深度学习算法的日均快递业务量预测模型[J]. 统计与决策, 2021, 37(13): 176-179.

[10] 韦军, 戚振强, 江玲庆, 等 . 人工智能在建设项目设计阶段的应用研究[J]. 建筑经济, 2026, 47(5): 18-25.

[11] Wei Jun, Qi Zhenqiang, Jiang Lingqing, et al. Research on the application of artificial intelligence in the design phase of construction projects[J]. Construction Economy, 2026, 47(5): 18-25.

[12] 段瀚, 宣勇, 曾思博. 以建筑机器人为新质生产力的工程价值链建构研究[J]. 建筑经济, 2026, 47(4): 16-25.

[13] Duan Han, Xuan Yong, Zeng Sibo. Research on the formation of engineering value chains with construction robots as new-qualitys[J]. Construction Economy, 2026, 47(4): 16-25.

[14] Kipf T N, Welling M. Semi-supervised classification with graph convolutional networks[C]// International Conference on Learning Representations. Toulon: OpenReview.net, 2017.

[15] 董学文, 石宇强, 田永政. 基于图神经网络的云制造服务推荐方法研究[J]. 工业工程, 2023, 26(5): 115-123.

[16] Dong Xuewen, Shi Yuqiang, Tian Yongzheng. A recommendation method for cloud manufacturing services based on graph neural networks[J]. Industrial Engineering Journal, 2023, 26(5): 115-123.

[17] Zheng C, Fan X, Wang C, et al. GMAN: a graph multi-attention network for traffic prediction[C]// Proceedings of the AAAI Conference on Artificial Intelligence: 01. New York: AAAI Press, 2020: 1234-1241.

[18] 汪维泰, 王晓强, 李雷孝, 等 . 时空图神经网络在交通流预测研究中的构建与应用综述[J]. 计算机工程与应用, 2024, 60(8): 31-45.

[19] Wang Weitai, Wang Xiaoqiang, Li Leixiao, et al. Review of construction and applications of spatio-temporal graph neural network in traffic flow prediction[J]. Computer Engineering and Applications, 2024, 60(8): 31-45.

[20] Yuan J, Zheng Y, Xie X. Discovering regions of different functions in a city using human mobility and POIs[C]// KDD'12: Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York: Association for Computing Machinery, 2012: 186-194.

[21] Yao Y, Li X, Liu X, et al. Sensing spatial distribution of urban land use by integrating points-of-interest and Google Word2Vec model[J]. International Journal of Geographical Information Science, 2017, 31(4): 825-848.

[22] Yan B, Janowicz K, Mai G, et al. From ITDL to Place2Vec: Reasoning about place type similarity and relatedness by learning embeddings from augmented spatial contexts[C]// SIGSPATIAL'17: Proceedings of the 25th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. New York: Association for Computing Machinery, 2017: 1-10.

[23] Huang W, Zhang D, Mai G, et al. Learning urban region representations with POIs and hierarchical graph infomax[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2023, 196: 134-145.

[24] Gao S, Janowicz K, Couclelis H. Extracting urban functional regions from points of interest and human activities on location-based social networks[J]. Transactions in GIS, 2017, 21(3): 446-467.

[25] Lee S, Kim J. A computerized operator support system to monitor the technical specifications[C]// Proceedings of the 29th European Safety and Reliability Conference(ESREL). Hannover: Research Publishing, 2019.

[26] Veličković P, Fedus W, Hamilton W L, et al. Deep graph infomax[C]// International Conference on Learning Representations. New Orleans: OpenReview.net, 2019.

[27] Wang T, Isola P. Understanding contrastive representation learning through alignment and uniformity on the hypersphere[C]// Proceedings of the 37th International Conference on Machine Learning. Virtual Event: PMLR, 2020: 9929-9939.

[28] Yu B, Yin H, Zhu Z. Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting[C]// Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence. Stockholm: International Joint Conferences on Artificial Intelligence Organization, 2018: 3634-3640.

[29] Oreshkin B N, Carpov D, Chapados N, et al. NBEATS: Neural basis expansion analysis for interpretable time series forecasting[C]// International Conference on Learning Representations. Addis Ababa: OpenReview.net, 2020.

基金资助

国家自然科学基金青年科学基金项目(72301170)

上海交通大学“新进青年教师启动计划”项目(23X010502006)