EURO 2025 Leeds
Abstract Submission

1472. An Integrated Framework for Population Synthesis and Activity Scheduling for Transportation Microsimulation

Invited abstract in session MD-59: Transport system and land use, stream Transportation.

Monday, 14:30-16:00
Room: Liberty 1.14

Authors (first author is the speaker)

1. Polina Maglevannaia
Center for Information Technologies, BioSense Institute
2. Branislav Pejak
Center for Information Technologies, BioSense Institute
3. Sanja Brdar
Center for Information Technologies, BioSense Institute
4. Oskar Marko
Center for Information Technologies, BioSense Institute
5. Vladimir Crnojević
BioSense Institute
6. Nikola Obrenović
BioSense Institute

Abstract

Urban transportation planning requires detailed insights into both socio-demographic composition and travel behavior. Yet, many regions rely on limited, aggregated, and outdated data, undermining accurate modeling. Our work addresses this challenge by proposing an integrated framework that generates high-fidelity synthetic households, individuals, and detailed daily travel schedules. Using an advanced MCMC-based one-step sampling approach, we synthesize populations that reproduce the joint distribution of key household and individual attributes, relying solely on aggregated census and survey records. The synthesized population serves as a critical input for an activity-based model, which simulates activities based on discrete choice models that account for individual preferences, time constraints, spatial, and modal characteristics. The integration of population synthesis and activity-based modeling overcomes data limitations and ensures consistency in capturing both marginal and joint distributions accurately reflecting the characteristics of real-life populations. By doing so, our framework enhances the behavioral realism of subsequent agent-based microsimulations and enables reliable evaluations of planned transportation infrastructure changes, such as traffic rerouting and pedestrian zone creation. Developed within the UDENE project, our approach offers a powerful, data-driven tool for sustainable urban planning and decision-making in data-constrained environments.

Keywords

Status: accepted

Back to the list of papers