Developped an optimized matching algorithm to generate a private carpooling program.
Used Julia : JuMP, GLPK, CPLEX, Gurobi
Project Description:
The project originated from a group of parents who annually create a carpooling schedule based on their children's timetables. This process, involving daily commutes from Saint-Cloud to Versailles, is time-consuming and tedious, often taking several days to complete. The objective is to develop an algorithm that can automate this task, reducing the burden on parents and saving valuable time. While initially designed for this specific case, the project is easily adaptable for multiple carpooling groups. Future goals include enhancing the user experience through a dedicated website. The project aims to generate a weekly carpooling schedule considering parental offers and children's requests, with the algorithm striving to maximize the number of fulfilled requests, optimize car occupancy, and prioritize matching drivers and passengers from the same family. Constraints include the number of available trips per child, which depends on the number of parental offers multiplied by the available seats in their vehicles, and the requirement for matching return trips based on parental offers.
Computer language : Julia (optimization algorithm)
Please read "Synthèse_Cassiopée n°51 _Elaboration d’un algorithme de matching pour générer un planning de covoiturage.pdf" for more information and details.
Please feel free to contact me if you need any help.