Logistics, Graphs, and Transformers: Towards improving Travel Time Estimation

TL;DR

This paper introduces TransTTE, a transformer-based model designed to improve travel time estimation by leveraging complex spatial and temporal data in logistics.

Contribution

The paper presents a novel transformer architecture tailored for travel time estimation, addressing the challenges of spatial-temporal data complexity in logistics.

Findings

TransTTE outperforms existing models in accuracy.

Transformer architecture effectively captures spatial-temporal dependencies.

Significant improvements over prior solutions in travel time prediction.

Abstract

The problem of travel time estimation is widely considered as the fundamental challenge of modern logistics. The complex nature of interconnections between spatial aspects of roads and temporal dynamics of ground transport still preserves an area to experiment with. However, the total volume of currently accumulated data encourages the construction of the learning models which have the perspective to significantly outperform earlier solutions. In order to address the problems of travel time estimation, we propose a new method based on transformer architecture - TransTTE.