Real-time dispatching of rebalancing tasks for a bike sharing network

Bike rebalancing involves the redistribution of bicycles among stations to mitigate potential system imbalances, addressing shortages or surpluses of bikes at individual stations. This task is done using trucks designed for moving bikes. The aim is to optimize the routing plan for these bike-moving trucks, deciding which stations to visit and the order to visit them. The plan also determines how many bikes to load and unload at each station. The primary objective is to prevent customer dissatisfaction caused by inefficiencies, such as stations running out of bikes as well as avoiding the situation where stations lack the capacity to drop the bike. The optimization framework seeks to minimize these undesirable outcomes while also reducing the operational cost of the system. This project focuses on the dynamic capacitated pickup and delivery vehicle routing problem. Given its NP-hard complexity, addressing this in real-time, particularly within a large-scale system, poses a substantial challenge necessitating advanced optimization techniques.

Faculty Supervisor:

Antoine Legrain

Student:

Partner:

LispLogics

Discipline:

Mathematics

Sector:

Professional, scientific and technical services

University:

Polytechnique Montréal

Program: