Development of a Dynamic Modeling Methodology for Urban Mobility with MATSim Using Dynamic Data

Offer Description

Keywords

Mobility simulation, MATSim/EQASim, mobility demand, OD matrix, data fusion, mobile phone data, public transport ticketing data

Context

MATSim is a widely-used activity-and-agent based simulation framework designed to analyze transportation systems and urban planning. The MATSim scenario [1-2] developed for the Île-de-France region utilizes open data and free software, traditionally relying on expensive survey data, which is updated every decade for mobility surveys and annually for household censuses. However, recent significant shifts in urban mobility patterns in Ile-de-France Region such as the increase in bicycle use from a 3% modal share in 2010 [3] to 11.2% in 2023 [4], the rise of remote work post-COVID-19 pandemic (ref.), and the introduction of Low Emission Zones (LEZ) (ref.),  challenge the model's ability to accurately capture current behaviors. These factors make the existing MATSim model, which typically depicts a standard day from past surveys, less effective for some mobility studies such as the development of new cycle paths or new public transport lines. The model struggles to capture both the dynamic and long-term shifts in mobility behavior. This presents a significant limitation for evaluating mobility policies.

To address these limitations, it is essential to integrate into the mobility demand generation process new and more recent sources of information such as dynamic data from GPS tracks, counting stations, mobile phone and public transport validation. These heterogeneous data sources provide fragmented but timely and localized insights into mobility trends, offering a contrast to the broader scope of traditional surveys. By leveraging these data sources, the MATSim simulation outputs can be enhanced and augmented through post-processing, enabling a more accurate representation of up-to-date travel behaviors. This approach facilitates also the representation of specific categories of days, rather than relying solely on a typical day model. It's important to note that while this approach complements the standard mobility surveys, it does not replace them entirely.

Nevertheless, aligning the augmented simulation outputs with the activity plans of the synthetic population originally generated from traditional surveys, poses a challenge. Several strategies can be employed to bridge this gap, such as modifying activity chains or incorporating additional demographic groups, like tourists. These adjustments aim to refine the synthetic population, ensuring it accurately mirrors individual behaviors and broader mobility trends as observed in the augmented simulation data.

In this context, this postdoctoral position aims to leverage the availability of dynamic mobility data, particularly from mobile phone data and public transport ticketing data to …. The primary objective is to analyze global mobility trends using mobile phone data [5-6], which offers extensive population coverage and comprehensive insights into mobility. A key component of this research involves comparing these insights found from mobile phone data with the most recent mobility survey. This comparison aims to evaluate the representativeness of mobile phone data against the conventional benchmarks established by traditional surveys during the same period, specifically regarding travel behaviors.

Following this, a deeper analysis of public transport mobility behaviors within the Île-de-France region for the same period will be conducted. The aim is to understand the connection between general mobility trends and public transport usage [7]. Additionally, the analysis aims to determine whether public transport mobility is sporadic or if it aligns with broader mobility trends. This analysis will enhance our understanding of mobility trends, both spatially and temporally, and guide the integration of these elements into synthetic population generation and mobility simulation.

Research topics

1. Analysis of Data Discrepancies: Conduct a comprehensive analysis of the differences between traditional mobility survey, mobile phone data/public transport ticketing data and existing simulation results to identify and understand the underlying causes of observed discrepancies in various data sources.
2. Global and Public Transport Mobility Trends: Investigate global mobility trends alongside public transport mobility trends, examining the connection between these two aspects to discern patterns and correlations.
3. Refinement of Simulation Outputs: Based on the insights gained from steps 1 and 2, utilize mobile phone data and public transport ticketing data to refine and augment simulation outputs. This refinement aims to enhance the representativeness of the simulated behaviors, ensuring they more accurately reflect contemporary mobility patterns.
4. Reconstruction of Individual Travel/Activity Plans: Develop a synthesized dataset by reconstructing individual travel/activity plans from the augmented simulation outputs. This dataset will serve as the foundation for creating a virtual mobility survey that mirrors current mobility behaviors.
5. Simulation Workflow and Validation: Execute the synthetic population generation and simulation workflow using data derived from the virtual mobility survey. This step is crucial for ensuring that the simulation outcomes align with empirical observations, validating the accuracy and relevance of the enhanced simulation model.

Principal responsibilities will encompass sophisticated analyses of extensive datasets, integration of diverse data sources, and calibration of the MATSim simulator. The postdoctoral researcher will investigate and potentially adapt existing data fusion methodologies [8], originally designed for bicycles, to public transportation systems. Additionally, he will engage in the reconstruction of individual mobility plans from augmented simulation outputs and the creation of virtual surveys, employing either established methodologies or developing innovative approaches.

Team

The candidate will join an experienced research team in mobility analysis and traffic estimation [9-12] with close connections to local and national stakeholders. He will furthermore be part of the Mob Sci-Dat Factory project, in partnership with CEREMA, IGN-ENSG, INRIA, and Université Gustave Eiffel, which aims to improve methods for collecting, processing, and analyzing heterogeneous mobility data.

For further information and to apply, please contact:

Guoxi Feng, Azise Oumar Diallo, Alexandre Chasse

Department of Control, Signal and System

IFP Energies Nouvelles, France

Email: guoxi.feng@ifpen.fr, azise-oumar.diallo@ifpen.fr, alexandre.chasse@ifpen.fr

References

[1] Hörl, S., & Balac, M. (2020). Open data travel demand synthesis for agent-based transport simulation: A case study of Paris and Île-de-France

[2] Feng, G., Jean, M., Chasse, A., & Hörl, S. (2021). Pre-calibration of a Discrete Choice Model and Evaluation of Cycling Mobility for Île-de-France. Procedia Computer Science, 184, 172-177.

[3] Île-de-France Mobilités, OMNIL, DRIEAT (2010). Enquête Globale Transport 2010.

[4] Institut Paris Région (2024). Enquête Mobilité par GPS.

[5] Hu, Y., Yang, C., Kagho, G. O., & Axhausen, K. W. (2022). Eqasim simulation using mobile phone signalling data: A case study of Shanghai, China. Arbeitsberichte Verkehrs-und Raumplanung, 1762.

[6] Matet, B., Côme, E., Furno, A., Hörl, S., & Oukhellou, L. (2023, September). Use of Origin-Destination data for calibration and spatialization of synthetic travel demand. In hEART'23. 11th Symposium of the European Association for Research in Transportation.

[7] Hu, Y., Yang, C., & Axhausen, K. W. (2023). Multi-modal travel simulation and travel behavior analysis: Case study in Shanghai. Arbeitsberichte Verkehrs-und Raumplanung, 1836.

[8] Feng, G., Diallo, A. O., & Chasse, A. Improving bike travel demand generation with dynamic data: an application to the Paris Metropolis.

[9] Michel, P., & Chasse, A. (2024). Aide à la décision au choix d’un véhicule. Les Techniques de l'Ingenieur, TRP904.

[10] Othman, B., De Nunzio, G., Laraki, M., & Sabiron, G. (2022, October). A novel approach to traffic flow estimation based on floating car data and road topography: Experimental validation in Lyon, France. In 2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC) (pp. 2571-2576). IEEE.

[11] Laraki, M., De Nunzio, G., & Othman, B. (2022, October). A large-scale and data-based road traffic flow estimation method leveraging topography information and population statistics. In 2022 IEEE 25th International Conference on Intelligent Transportation Systems (ITSC) (pp. 2343-2349). IEEE.

[12] Carneiro, S. A., Chierchia, G., Charléty, J., Chataignon, A., & Najman, L. (2023, June). Swmlp: Shared weight multilayer perceptron for car trajectory speed prediction using road topographical features. In 2023 8th International Conference on Models and Technologies for Intelligent Transportation Systems (MT-ITS) (pp. 1-6). IEEE.

Where to applyE-mailalexandre.chasse@ifpen.fr

Requirements

Research Field:

Mathematics » Statistics or Engineering

Education Level:

PhD or equivalent

Skills/Qualifications

• Education: PhD in Computer Science, Statistics, Transport Engineering, or a closely related field.
• Technical Skills:
• Experience with multi-agents simulation tools
• Proficiency in data analysis techniques and advanced programming skills, particularly in Python/SQL.
• Practical Experience:
• Proven experience in applying mobility simulation techniques to real-world case studies, with a focus on analyzing and interpreting complex mobility patterns.

Specific Requirements

Level C1 in English. Willingness to learn French.

Languages: ENGLISH Level: Good

Work Location(s)

IFP Energies nouvelles, 1 et 4 avenue de Bois Préau, 92852 Rueil-Malmaison, France