Multi-Modal Transport Flow Allocation Model

MIRACA (Multi-hazard Infrastructure Risk Assessment for Climate Adaptation) is a research project developing an evidence-based decision-support toolkit that responds to real-world infrastructure risk and adaptation challenges.

This project has received funding from the European Union’s Horizon Europe research programme under grant agreement No 101004174.

Deliverable 2.4: Creating transport flows on the network

Overview

This repository provides the integrated multi-modal transport flow allocation model developed within the MIRACA project. The model represents passenger and freight flows across the European transport system, including maritime, air, rail, road, and inland waterway (IWW) networks.

The transport modelling framework builds upon previously published methodologies for flow allocation and spatial interaction modelling, integrating Eurostat Supply–Use Tables, facility-scale industrial datasets (EPRTS), population distributions (GHSL), and multi-modal transport network representations aligned with the TEN-T corridors. Freight flows are statistically downscaled based on the spatial distribution of industrial production, while passenger flows are allocated using population cluster nodes.

Model implementation

The repository includes the full implementation of the following components:

Statistical downscaling of commodity trade flows to industrial production locations

Allocation of passenger flows to population cluster nodes

Multi-modal modal split estimation based on transport hub accessibility, connectivity, and capacity

Flow allocation for maritime, air, rail, road, and inland waterway (IWW) networks

Identification of critical transport lifelines based on flow intensity and network topology

At a high level, the allocation of passenger and freight flows follows the same three-step process:(1) the creation of location-specific data describing sources (origins) and sinks (destinations) where trips start and end;(2) the estimation of movement volumes (expressed as passenger trips and freight tonnage) between origin-destination (OD) pairs, resulting in OD matrices; and(3) the assignment of these OD matrices to their respective transport modes based on network availability, capacity constraints, and transport cost functions.

The Python Jupyter notebooks provide a practical framework for analysing and visualising multimodal transport flows, with a specific focus on air, maritime, inland waterways, road, and rail networks. The notebooks include the required data inputs and executable workflows to estimate transport flows for individual countries as case studies. As the same codebase is used throughout, the framework is fully scalable and directly applicable at the pan-European level.

Outputs and results

The model produces transport flow results that can be accessed either as mode-specific data frames or as aggregated, joined outputs. These include:

Passenger and freight flow volumes per node and edge

Inbound and outbound flows for ports and airports

Modal flow distributions across each European transport network

Identification of high-intensity transport corridors

Visual outputs are generated through dedicated plotting routines, including bubble maps for maritime and air transport, and edge-weighted network representations for rail, road, and inland waterways.

Integration within MIRACA

This transport flow model forms a core component of the MIRACA analytical framework and complements other sectoral models developed within the project. It supports subsequent analyses of network and systemic risk, providing a consistent representation of transport dependencies across Europe.

The implementation presented here constitutes Deliverable 2.4, focusing on the generation and allocation of transport flows within the European multi-modal network, and builds upon earlier MIRACA deliverables addressing hazard modelling and infrastructure representation.

Accessibility

The model is accessible through the MIRACA GitHub repository, which contains the main program, supporting scripts, and documentation. The repository includes a Jupyter Notebook ((https://github.com/miracaEU/Transport-Flows-Allocation.git) demonstrating a study-case application of the model and showcasing key outputs and visualizations.

Acknowledgments

This deliverable builds upon existing literature on transport modelling, spatial interaction models, and infrastructure risk analysis, in addition to the sources already cited in previous MIRACA deliverables.