Open data travel demand synthesis for agent-based transport simulation: A case study of Paris and Île-de-France

Synthetic populations of travellers and their detailed mobility behaviour are an important basis for agent-based transport simulations, which are increasingly used in transport planning and research today. While previous applications of such simulations put focus on analyzing policies and novel modes of transport, less importance is put on the aspects of data collection, processing and validation to create the disaggregate travel demand. This paper proposes an open-source and extendable pipeline for travel demand synthesis for Île-de-France, which establishes a clear path from raw data to the final demand consisting of households, persons, and trips with their respective attributes. The travel demand synthesis is based on open and publicly accessible data and can be easily transferable to other regions in France. By basing the process on open software and open data, the paper establishes a baseline not only for generating a fully replicable travel demand, but also enables researchers to perform replicable transport simulations. While the proposed pipeline, which is runnable by any reader, is based on rather straightforward and data-driven algorithms, pathways are provided for extending the process. Additionally, methods for quality assessment of synthetic populations and travel demand are proposed and discussed