GMDNet: A Graph-Based Mixture Density Network for Estimating Packages’ Multimodal Travel Time Distribution

Authors

  • Xiaowei Mao School of Computer and Information Technology, Beijing Jiaotong University Artificial Intelligence Department, Cainiao Network
  • Huaiyu Wan School of Computer and Information Technology, Beijing Jiaotong University Beijing Key Laboratory of Traffic Data Analysis and Mining
  • Haomin Wen School of Computer and Information Technology, Beijing Jiaotong University Artificial Intelligence Department, Cainiao Network
  • Fan Wu Artificial Intelligence Department, Cainiao Network
  • Jianbin Zheng Artificial Intelligence Department, Cainiao Network
  • Yuting Qiang Artificial Intelligence Department, Cainiao Network
  • Shengnan Guo School of Computer and Information Technology, Beijing Jiaotong University Beijing Key Laboratory of Traffic Data Analysis and Mining
  • Lixia Wu Artificial Intelligence Department, Cainiao Network
  • Haoyuan Hu Artificial Intelligence Department, Cainiao Network
  • Youfang Lin School of Computer and Information Technology, Beijing Jiaotong University Beijing Key Laboratory of Traffic Data Analysis and Mining

DOI:

https://doi.org/10.1609/aaai.v37i4.25578

Keywords:

DMKM: Mining of Spatial, Temporal or Spatio-Temporal Data, APP: Transportation

Abstract

In the logistics network, accurately estimating packages' Travel Time Distribution (TTD) given the routes greatly benefits both consumers and platforms. Although recent works perform well in predicting an expected time or a time distribution in a road network, they could not be well applied to estimate TTD in logistics networks. Because TTD prediction in the logistics network requires modeling packages' multimodal TTD (MTTD, i.e., there can be more than one likely output with a given input) while leveraging the complex correlations in the logistics network. To this end, this work opens appealing research opportunities in studying MTTD learning conditioned on graph-structure data by investigating packages' travel time distribution in the logistics network. We propose a Graph-based Mixture Density Network, named GMDNet, which takes the benefits of both graph neural network and mixture density network for estimating MTTD conditioned on graph-structure data (i.e., the logistics network). Furthermore, we adopt the Expectation-Maximization (EM) framework in the training process to guarantee local convergence and thus obtain more stable results than gradient descent. Extensive experiments on two real-world datasets demonstrate the superiority of our proposed model.

Downloads

Published

2023-06-26

How to Cite

Mao, X., Wan, H., Wen, H., Wu, F., Zheng, J., Qiang, Y., Guo, S., Wu, L., Hu, H., & Lin, Y. (2023). GMDNet: A Graph-Based Mixture Density Network for Estimating Packages’ Multimodal Travel Time Distribution. Proceedings of the AAAI Conference on Artificial Intelligence, 37(4), 4561-4568. https://doi.org/10.1609/aaai.v37i4.25578

Issue

Section

AAAI Technical Track on Data Mining and Knowledge Management