[AUTRES] SIG et Multi-Agent Modeling - Compiegne (60)

Participez à la sélection du programme !

----- French version below -----
Internship Offer – "Analysis of Access to Everyday Life Resources Using Multi-Agent Modeling"
Hosting Unit: AVENUES Laboratory
Supervisors:

• Berk Celik, Associate Professor in Electrical Engineering
• Maxime Hachette, Associate Professor in Urban Planning

Keywords:
Accessibility ergonomics, sustainable mobility, multi-agent modeling, socio-spatial inequalities, decision support

Introduction
Since the 1980s, many European cities have implemented policies to reduce the negative impacts of transport and mobility, particularly those caused by private cars. In the Eurométropole de Strasbourg, measures have been introduced to limit car use while promoting sustainable transportation. These measures include the development of public transportation, the promotion of shared and active modes (walking, cycling), and urban planning focused on a "proximity metropolis" model to support sustainable mobility.
To analyze how cities enable their residents to access essential everyday resources efficiently, comfortably, and affordably (in terms of cost, effort, risk, emissions, energy, and time) while using sustainable transport modes, spatial ergonomics—which studies a territory’s ability to meet its inhabitants' needs at a lower cost—and more specifically, accessibility ergonomics, which approaches the issue from a mobility perspective, provide a relevant analytical framework.
Activity-based transport models are used to simulate travelers’ behaviors and interactions, overcoming some limitations of aggregated four-step transport models. They are already employed to assess accessibility through a trip completion rate, measuring the proportion of trips below a given threshold (time or distance). However, existing approaches do not account for key aspects of accessibility ergonomics (such as safety and convenience).
This internship will contribute to the development of a multi-agent model designed to predict access to everyday resources. The intern will help create a synthetic population, a representation of travel demand in the Eurometropolis of Strasbourg, and the overall model.

Tasks
The intern will contribute significantly to the following tasks:

• Literature review on energy and mobility modeling
• Data collection and implementation of a multi-agent model

Methodology
To simulate access to everyday resources, the following steps are anticipated:

1. Data collection and analysis: Gathering relevant datasets, including census data, statistical zoning systems, and household travel surveys. Given Strasbourg's border location, the inclusion of German data will be considered.
2. Synthetic population generation: Adapting an existing pipeline to generate a synthetic population.
3. Evaluation of the synthetic population: Using the synthetic population in a multi-agent travel model to assess its accuracy, including comparisons with traffic data.

Through this methodology, the internship will also contribute to optimizing electric vehicle charging infrastructure by integrating GIS tools and Python programming for sustainable urban mobility solutions.

Research Plan
Task Duration

1. Task I: Literature review and current trends - 1 month
2. Task II: Evaluation of methodology and case study development - 1 month
3. Task III: Data collection and processing - 2 months
4. Task IV: Synthetic population evaluation - 1 month
5. Task V: Internship report preparation - 1 month

Desired Profile

• Education Level: Final year of an engineering or master’s degree program
• Basic knowledge of Python (familiarity with NumPy and Pandas for data processing is a plus)
• Experience in transport modeling, particularly multi-agent models, is appreciated. Experience with multi-agent modeling in other fields or theoretical knowledge in demand modeling and macroscopic/mesoscopic assignment will also be considered.
• Languages: Good proficiency in English for international collaboration; knowledge of French is a plus.

Internship Details

• Start Date: Early 2025
• Duration: 6 months
• Hosting Institution:
  Laboratory: AVENUES Lab / UTC (Université de Technologie de Compiègne)
  Address: Rue du Docteur Schweitzer, Centre Pierre Guillaumat 2, 60203 Compiègne, France
  Website: https://avenues.utc.fr/

Application Process
Candidates should first contact B. Celik, M. Hachette, and C. Hacquard as soon as possible. Then, they should send their CV and cover letter by February 21, 2025. Although not mandatory, applications including a sample of previous work will be highly valued.
Contacts:

• Berk Celik – Email: berk.celik@utc.fr
• Maxime Hachette – Email: maxime.hachette@utc.fr
• Clément Hacquard – Email: clement.hacquard@utc.fr

Maxime Hachette
Associate Professor / Maître de Conférences