Express Time Prediction Method Based on Multi-Task Learning

doi:10.19678/j.issn.1000-3428.0061098

Abstract

Abstract: Delivery time prediction(i.e., predicting package arrival time at any time) is important to logistics service providers.Accurate prediction of the delivery time provides customers with more prompt services and alleviates anxiety. It is also beneficial to the route planning by couriers for improved delivery efficiency.In real scenarios, however, accurate delivery time prediction is marred with multiple destinations, multiple factors, and dynamics challenges.In this paper, relying on the historical spatio-temporal trajectories of couriers, a Multi-Task model for Delivery Time prediction Network(MTDTN) is proposed to predict the package delivery time.MTDTN leverages external factors that may affect the delivery time and utilizes the geographic information encoder, convolution operation, and the Bidirectional Long Short-Term Memory(Bi-LSTM) to capture the spatio-temporal information in the trajectories.Moreover, multi-task learning is used to simultaneously predict both the delivery time and the delivery sequence.The model performance is enhanced by introducing the delivery sequence prediction as an auxiliary task.Experimental results on real data sets show that, compared with the optimal DeepETA model in the benchmark method, Mean Absolute Error(MAE) and Mean Absolute Percentage Error(MAPE) of this model are reduced by 16.11% and 12.88% respectively.

Key words: delivery time prediction, spatial-temporal trajectory, Convolutional Neural Network(CNN), Bidirectional Long Short-Term Memory(Bi-LSTM), multi-task learning

摘要： 快件送货时间预测（即在任何时间预测包裹送达的到达时间）是物流领域中最重要的服务之一。准确地预测快件送达时间可以为用户提供更准时的服务，缓解客户的等待焦虑，提升用户体验，且有利于快递员的路径规划，从而提高派送效率。然而在快递派送场景下，多因素、动态性及多目的地等特征给快件投递准确预测送达时间带来巨大挑战。提出一种基于多任务学习的模型MTDTN，从快递员的大量历史时空轨迹中预测快件送达时间。MTDTN建模多种影响送达时间的外部因素，利用地理信息编码、卷积操作以及双向长短时记忆网络来捕获派送行为的时空关系，并运用多任务学习框架，引入顺序预测的辅助任务与送达时间预测的主任务，提高模型预测性能。在真实数据集上的实验结果表明，与基准方法中最优的DeepETA模型相比，该模型的平均绝对误差与平均绝对百分比误差分别降低了16.11%和12.88%，模型效果明显提升。

关键词: 送达时间预测, 时空轨迹, 卷积神经网络, 双向长短时记忆, 多任务学习

CLC Number:

TP391

WANG Qiang, LIN Youfang, WAN Huaiyu. Express Time Prediction Method Based on Multi-Task Learning[J]. Computer Engineering, 2022, 48(4): 314-320.

王强, 林友芳, 万怀宇. 基于多任务学习的快件送达时间预测方法[J]. 计算机工程, 2022, 48(4): 314-320.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0061098

https://www.ecice06.com/EN/Y2022/V48/I4/314

Figures/Tables 3

References

[1] 王岳含, 曾军山.快递业自赋能现状及发展方向[J].中国国情国力, 2020(1):56-60. WANG Y H, ZENG J S.The status and development direction of self-empowerment in express delivery industry[J].China National Conditions and Strength, 2020(1):56-60.(in Chinese)
[2] WU F, WU L X.DeepETA:a spatial-temporal sequential neural network model for estimating time of arrival in package delivery system[J].Artificial Intelligence, 2019, 33(1):774-781.
[3] ULMER M W, THOMAS B W.Enough waiting for the cable guy-estimating arrival times for service vehicle routing[J].Transportation Science, 2019, 53(3):897-916.
[4] 闫皎洁, 张锲石, 胡希平.基于强化学习的路径规划技术综述[J].计算机工程, 2021, 47(10):16-25. YAN J J, ZHANG Q S, HU X P.Review of path planning techniques based on reinforcement learning[J].Computer Engineering, 2021, 47(10):16-25.(in Chinese)
[5] HONG H T, LIN Y C, YANG X Q, et al.HetETA:heterogeneous information network embedding for estimating time of arrival[C]//Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York, USA:ACM Press, 2020:2444-2454.
[6] SUN Y W, WANG Y L, FU K, et al.FMA-ETA:estimating travel time entirely based on FFN with attention[EB/OL].[2021-02-10].https://arxiv.org/abs/2006.04077v1.
[7] LECUN Y, BENGIO Y, HINTON G.Deep learning[J].Nature, 2015, 521(7553):436-444.
[8] HOCHREITER S, SCHMIDHUBER J.Long short-term memory[J].Neural Computation, 1997, 9(8):1735-1780.
[9] PARK C, LEE C, BAHNG H, et al.STGRAT:a spatio-temporal graph attention network for traffic forecasting[EB/OL].[2021-02-10].https://arxiv.org/abs/1911.13181v2.
[10] YAO H X, TANG X F, WEI H, et al.Revisiting spatial-temporal similarity:a deep learning framework for traffic prediction[J].Artificial Intelligence, 2019, 33(1):5668-5675.
[11] LI K, YAN X J, TAN X H, et al.Predicting taxi demand based on 3d convolutional neural network and multi-task learning[J].Remote Sensing, 2019, 11(11):1265.
[12] WANG Z, FU K, YE J P.Learning to estimate the travel time[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York, USA:ACM Press, 2018:858-866.
[13] LEE W C, SI W P, CHEN L J, et al.HTTP:a new framework for bus travel time prediction based on historical trajectories[C]//Proceedings of ACM International Symposium on Advances in Geographic Information Systems.New York, USA:ACM Press, 2012:279-288.
[14] WENG J C, WANG C, HUANG H N, et al.Real-time bus travel speed estimation model based on bus GPS data[J].Advances in Mechanical Engineering, 2016, 8(11):1687-1697.
[15] WANG Y L, ZHENG Y, XUE Y X.Travel time estimation of a path using sparse trajectories[C]//Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York, USA:ACM Press, 2014:25-34.
[16] ZHANG H Y, WU H, SUN W W, et al.DeepTravel:a neural network based travel time estimation model with auxiliary supervision[C]//Proceedings of the 27th IEEE International Joint Conference on Artificial Intelligence.Washington D.C., USA:IEEE Press, 2018:3245-3256.
[17] SCHUSTER M, PALIWAL K K.Bidirectional recurrent neural networks[J].IEEE Transactions on Signal Processing, 1997, 45(11):2673-2681.
[18] WANG D, ZHANG J B, CAO W, et al.When will you arrive?Estimating travel time based on deep neural networks[C]//Proceedings of AAAI Conference on Artificial Intelligence.[S.1.]:AAAI Press, 2018:2366-2378.
[19] LI Y G, FU K, WANG Z, et al.Multi-task representation learning for travel time estimation[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York, USA:ACM Press, 2018:1695-1704.
[20] GAL Y, GHAHRAMANI Z.A theoretically grounded application of dropout in recurrent neural networks[EB/OL].[2021-02-10].https://arxiv.org/pdf/1512.05287.pdf.
[21] ZHANG J B, ZHENG Y, QI D K, et al.DNN-based prediction model for spatio-temporal data[C]//Proceedings of the 24th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems.New York, USA:ACM Press, 2016:1-4.

Please choose a citation manager

Content to export