Le 22/11/2023 par drivreau :
COMPUTER SCIENCE | AI & OR INTERNSHIP
Fueling Efficiency: Harnessing Machine Learning and Operations Research for Optimal Technician Tour Scheduling in Gas Operations
Keywords: Machine Learning, Operations Research, Vehicle Routing,
PROJECT MOTIVATION & GOALS: In the high-pressure gas transportation sector, the main purpose of gas transmission companies is to operate buried pipelines to transport gas from suppliers to consumers. They must ensure the continuity of transmission of gas while safeguarding the safety of individuals and property. To this end, they conduct a regular monitoring of the pipeline network to detect potential anomalies and engage in preventive and corrective maintenance of the infrastructure. From the operational point of view, surveillance and maintenance activities are typically carried out by various teams of technicians who undertake tours utilizing aerial means (aircraft), road vehicles (cars), and/or ground-based methods (on foot). The substantial growth in biomethane production (gas produced from organic waste) and the imminent arrival of hydrogen will have an impact on the operation and maintenance of these facilities. Furthermore, in order to ensure the long-term competitiveness of gas, new surveillance methods such as drones and satellites are being considered. In this context, the objective of this internship is to develop methods that combine machine learning and operations research to optimize the scheduling of technician tours within a French company and assess the impact of forthcoming changes.
CONTEXT: The internship will be funded by the University of Angers. The intern will be based either at the Interuniversity Research Center on Enterprise Networks, Logistics, and Transportation – CIRRELT (https://www.cirrelt.ca) in the city of Montreal (Canada), or at the Laboratoire Angevin de Recherche en Ingénierie des Systèmes - LARIS (https://laris.univ-angers.fr ) in the city of Angers (France). The internship will be supervised by Professors Martin Cousineau (HEC Montréal), Christelle Guéret (University of Angers) and David Rivreau (Université Catholique de l'Ouest). The intern is also expected to collaborate with Nadia Ghernaout (PhD student). The intern will approximately receive a scholarship of 550 euros per month, plus reimbursement of travel (airfare) and accommodation expenses up to 2,000 euros if a stay in Canada is planned. The duration of the internship is 4 to 6 months, depending on the intern’s availability. The start date is flexible and can be fit to the candidate's earliest convenience. There is also a possibility to continue to a PhD after the internship.
DESIRED QUALIFICATIONS: The ideal applicant will be fluent in French or English and have experience with machine learning in Python (e.g., scikit-learn, PyTorch, TensorFlow) and operations research solution methods (e.g., integer programming, metaheuristics).
CONTACT: Interested applicants should email Martin Cousineau (martin.cousineau@hec.ca) with the following attachments: a brief letter of motivation, an up-to-date CV, transcripts for the last two academic years, and the name and contact information of 2 references. This position will remain open until filled.
