Temporal distribution of travel demand provides valuable insights into the planning and operation of transport systems. As a key input to dynamic traffic assignment (DTA) models, estimation of time-dependent origin-destination (TDOD) traffic demand matrices across the modelled network gained attention in the 1980s, with significant advancements in methods and techniques since then. However, the strong reliance on observed traffic counts has long been recognised as a limitation of these approaches. In the travel demand modelling (TDM) domain, the time-dependency of travel demand has been associated with travellers' characteristics, typically implemented through theory-based choice modelling (CM) methods informed by stated or revealed preference datasets. CM's reliance on preference datasets introduces its own limitations, e.g. the high cost of conducting reliable surveys, which constrain its broader applicability for predicting the temporal distribution of travel demand. This paper demonstrates the successful application of machine learning to predict the temporal distribution of origin-destination (OD) traffic demand using TDM data, including sociodemographic and land use information of the origin and destination zones as well as OD level network statistics, e.g. travel time. By incorporating the TDM data and available count-based TDOD estimates, we construct a combined dataset, which is partitioned into training, validation and test sets to train and evaluate the machine learning models. Results show that the trained models accurately predict the temporal distribution of origin-destination traffic demand. Our approach effectively addresses the limitations associated with count-based estimation and theory-driven choice modelling approaches.
Predicting the temporal distribution of origin-destination traffic demand using machine learning
2025
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
Using Google’s Passive Data and Machine Learning for Origin-Destination Demand Estimation
Transportation Research Record | 2018
|Dynamic origin–destination demand flow estimation under congested traffic conditions
Online Contents | 2013
|Contextualized Spatial-Temporal Network for Taxi Origin-Destination Demand Prediction
ArXiv | 2019
|