진학프로석·박사 채용 접수 플랫폼

Context

The transport sector accounts for nearly 30% of French national greenhouse gas (GHG) emissions, with around 94% originating from road transport alone [1]. This dominance is explained by the central role of road transport in France, which represents 88% of freight and 85% of passenger transport [2]. Another key factor is the sector’s strong dependence on fossil fuels, as most vehicles today rely on internal combustion engines powered by petroleum-based fuels. Road transport thus faces two urgent and interrelated challenges: reducing air pollution and decarbonizing to limit its contribution to global warming.

To address these issues, new legislative frameworks have been introduced. The Energy Transition Law for Green Growth (2015) and the Mobility Orientation Law (2019) promote measures such as the Crit’Air certification and the creation of Low Emission Zones (ZFE-m), gradually restricting access to highly polluted urban areas for the most polluting vehicles. Moreover, both national and European regulations—such as the European Regulation 2023/2772 and Article L.1431-3 of the French Transport Code—now require transport operators to publish their GHG emission data, reinforcing the demand for transparency.

Simultaneously, the energy and vehicle markets have undergone significant transformation. Where once internal combustion vehicles dominated, today’s market includes a wide range of alternative powertrains: hybrid electric vehicles (HEVs), battery electric vehicles (BEVs), and hydrogen fuel cell vehicles (FCEVs). Renewable fuels such as biodiesel (B100), HVO, and biogas have also emerged, offering new options for decarbonization, especially for heavy-duty and utility vehicles.

In this context, vehicle replacement decisions have become highly complex, influenced by evolving technologies, economic constraints, environmental goals, and legal obligations [3]. The tool Verdir ma Flotte (VMF, https://verdirmaflotte.fr/), developed by IFPEN with the support of ADEME [4], serves as a decision-support and educational platform. It enables users to compare TCO and GHG emissions across a comprehensive range of fuels and powertrains—fossil fuels (diesel, gasoline, natural gas), biofuels (B100, HVO, bioGNC), and electrified systems (HEV, BEV, FCEV)—for multiple vehicle types (light vehicles, vans, buses, coaches, and trucks). Users can customize data to reflect specific use cases, supporting more tailored and informed assessments.

While VMF supports unit-level decision-making, it does not address the strategic optimization of fleet-wide renewal plans under uncertainty—a critical need for organizations aiming to decarbonize cost-effectively and in compliance with future regulations. Although VMF offers a useful comparative framework for these choices, it does not enable a strategic, fleet-wide approach. In particular, some cost elements are shared across multiple vehicles and must be treated transversally—for instance, the CAPEX of a new biogas tank or the OPEX of specialized hydrogen maintenance affect the entire fleet, not just a single vehicle.

Objective

This postdoctoral research aims to develop and implement a robust optimization framework for strategic fleet renewal planning under deep uncertainty. The overall goal is to support decision-makers in determining when and how to renew vehicles—selecting the most appropriate year and energy type—while balancing GHG reduction objectives with investment (CAPEX) and operational (OPEX) constraints.

Tasks

1. Modeling and analysis of uncertainties
The first phase of the research will focus on identifying, structuring, and integrating major sources of long-term uncertainty that influence strategic fleet renewal decisions. This phase will include:

• Comparative analysis of prospective energy cost scenarios by reviewing and contrasting key national and international reference scenarios. In particular, bounding scenarios will be defined based on sources such as the Stratégie Nationale Bas Carbone[5], ADEME’s Transitions 2050 [6], the Alliance nationale de coordination de la recherche pour l’énergie [7], and the IEA’s World Energy Outlook [8]. These scenarios will serve to characterize a plausible range of techno-economic futures relevant for vehicle fleet planning.
• Definition and selection of bounding scenarios to structure the uncertainty space and guide the development of robust optimization strategies. This will involve identifying key uncertain variables (e.g., energy prices, technology availability, regulatory constraints), and defining a set of bounding or enveloping scenarios that span a wide range of plausible futures. These scenarios will be used to delineate the uncertainty space, capturing both optimistic and conservative trajectories of the energy transition.
• Sensitivity analysis to quantify the influence of key uncertain parameters on total cost of ownership (TCO), GHG emissions, and strategic fleet decisions. Special attention will be given to fleet-level effects such as pooled CAPEX (e.g., infrastructure for alternative fuels) and OPEX (e.g., maintenance costs tied to new technologies), which are shared across vehicle subsets and introduce structural dependencies in the optimization.

2. Robust optimization of fleet renewal strategies
The second phase will consist in designing and comparing optimization methods that remain effective across a wide range of plausible futures:

• Formulating a fleet-wide optimization model that integrates both vehicle-level decisions and shared fleet-level costs (e.g., infrastructure investments for refueling or maintenance equipment). The model will optimize key control variables, such as the replacement year and the energy and technology choice for each vehicle, subject to operational constraints, decarbonization targets, and cost considerations.
• Define objective functions and robustness criteria to guide decision-making, focusing on minimizing GHG emissions, capital (CAPEX) and operating costs (OPEX), while ensuring strong performance across a range of future scenarios. This involves managing worst-case outcomes, reducing variability, and limiting the risk of exceeding budgets or emission targets.
• Develop and evaluate advanced optimization approaches, focusing on selecting and adapting the most suitable techniques to minimize costs, environmental impact, and uncertainty. The goal is to enable robust and adaptive decision-making for complex fleet renewal planning under uncertain future scenarios.
• Applying and validating the optimization framework on realistic fleet configurations, using and extending the dataset provided by the Verdir ma Flotte (VMF) tool. This will ensure practical relevance and demonstrate the added value of moving from a comparative tool to a strategic planning model.

This work will build on recent studies in low-carbon fleet renewal and optimization under uncertainty, including approaches based on dynamic programming [9], fuel price and demand scenarios [10], and simulation-based methods [11].

This research will provide both theoretical and practical contributions to robust decision-making in fleet decarbonization, supporting the shift from pedagogical tools to actionable, fleet-wide planning instruments in a rapidly evolving regulatory and technological landscape.

Prerequisites

Education

• PhD in operations research, applied mathematics, industrial engineering, or a related field, with a strong focus on optimization.

Technical Skills

• Proficiency in programming, preferably in Python or a similar language commonly used in optimization.
• Familiarity with mathematical modeling tools and solvers (e.g., Pyomo, Gurobi, CPLEX).
• Knowledge of uncertainty modeling, robust or stochastic optimization is a strong asset.

Practical Experience

• Experience applying optimization methods to real-world problems in sectors such as energy, mobility, or logistics.
• Understanding of the energy system and/or transport and logistics networks is a plus.
• Exposure to techno-economic modeling or decision-support tools is appreciated.
• Motivation for applied research at the interface of modeling, data, and decision-making in public or industrial contexts.

Bibliography

[1] Commissariat général au développement durable, "Les émissions de gaz à effet de serre du secteur des transports," 25 02 2021. [Online]. Available: https://www.notre-environnement.gouv.fr/themes/climat/les-emissions-de-…. [Accessed 07 02 2025].

[2] Ministère de de la transition écologique, "Chiffres clés des transports - Édition 2022 - Mars 2022 - Transport terrestre de marchandises," 2022. [Online]. Available: https://www.statistiques.developpement-durable.gouv.fr/edition-numeriqu…. [Accessed 07 02 2025].

[3] P. Michel and A. Chasse, "Decision Support to Vehicle Choice," Techniques de l'ingénieur, 2024.

[4] P. Michel and A. Chasse, "Verdir ma flotte, un outil d'aide à la décision pour renouveler son véhicule," 2023.

[5]

Ministère de la Transition écologique, "Stratégie nationale bas-carbone (SNBC)," 24 10 2024. [Online]. Available: https://www.ecologie.gouv.fr/politiques-publiques/strategie-nationale-b…. [Accessed 13 02 2025].

[6] ADEME, "Prospective - Transitions 2050 - Rapport," 2021.

[7] ANCRE, "Scénarios de décarbonation du secteur transport en France et leurs impacts sur la biomasse, l’hydrogène et l’électricité," 2023.

[8] International Energy Agency, "World Energy Outlook 2024," Paris, 2024.

[9] J. Winkelmann, S. Spinler and T. Neukirchen, "Green transport fleet renewal using approximate dynamic programming: A case study in German heavy-duty road transportation," Transportation Research Part E: Logistics and Transportation Review, vol. 186, p. 103547, 2024.

[10] H. H. Turan, S. Elsawah and M. J. Ryan, "A long-term fleet renewal problem under uncertainty: A simulation-based optimization approach," Expert Systems with Applications, vol. 145, p. 113158, 2020.

[11] S. Zheng and S. Chen, "Fleet replacement decisions under demand and fuel price uncertainties," Transportation Research Part D: Transport and Environment, vol. 60, pp. 153-173, 2018.