dynamic-OD-estimation-railway-network

Code associated to the paper "Estimation of dynamic Origin-Destination matrices in a railway transportation network integrating ticket sales and passenger count data" authored by Greta Galliani, Piercesare Secchi, and Francesca Ieva

Overview

This repository contains the code and resources associated with the research paper titled Estimation of dynamic Origin-Destination matrices in a railway transportation network integrating ticket sales and passenger count data authored by Greta Galliani, Piercesare Secchi, and Francesca Ieva.

In particular:

• Script 1_estimation_trenord_od_matrices.R develops the code needed to estimate weekly OD matrices in the Trenord network, given ticket and counter data.
• Script 2_dynamics_network_analysis.R applies some techniques of network and functional data analysis to perform anomaly detection in the temporal weighted directed network induced by the Trenord dynamic OD matrices.

Moreover, script data_visualization.R produces all the plots needed to interpret the analyses' results.

The code is developed in R (version 4.2.1).

Data and Resources

Due to confidentiality agreements, the Trenord data about tickets sold and passenger counts in 2022 is unavailable. We produced two artificial datasets, ticket.csv and train.csv for demonstrative purposes, which, however, have no relationship with the actual data about railway movements in the Trenord railway network in 2022.

Citation

If you use this code or data in your work, please consider citing the original paper authored by Greta Galliani, Piercesare Secchi, and Francesca Ieva.

For further details about the research and its findings, please refer to the associated paper.