EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

4078. A bicycle rebalancing model to minimize cost of operations and unmet demand in a docked network

Invited abstract in session TA-56: Methods and models for sustainable transport solutions, stream Transportation.

Tuesday, 8:30-10:00
Room: S04 (building: 101)

Authors (first author is the speaker)

1. Aliki Pouliasi
Rural and Surveying Engineering, National Technical University of Athens
2. Amalia Ι. Nikolopoulou
Civil Enginering, National Technical University of Athens
3. Konstantinos Gkiotsalitis
Civil Engineering, NTUA
4. Konstantinos Kepaptsoglou
Rural and Surveying Engineering, National Technical University of Athens

Abstract

Bicycle sharing systems have the potential to significantly alleviate traffic congestion and minimize demand for parking spaces in urban centers. One critical factor influencing the success of a bicycle sharing system is the effectiveness of rebalancing operations. These operations involve restoring the number of bicycles at each station to its expected demand through pickup and delivery activities performed by trucks. The Static Docked Bicycle Rebalancing Problem (SBRP) focuses on determining a cost-effective sequence of visited stations for a truck, along with the corresponding quantity of bicycles to be picked up or delivered at each station. Deviating from past studies, we propose a new formulation for the SBRP problem which minimizes the cost of rebalancing operations, while factoring in the cost of unmet passenger demand. The developed mathematical model is a mixed-integer nonlinear program, which is reformulated as a mixed-integer linear program. Then, our study introduces an exact branch-and-cut algorithm for addressing the SBRP and presents the results of computational tests. Lazy constraints and valid inequalities are introduced to improve performance and reduce computational time. The experimental findings demonstrate that we can solve to global optimality instances with up to 20 stations. A construction heuristic (nearest neighbor) is also introduced to provide a warm start to the branch-and-cut search, resulting in a more effective search of the solution space.

Keywords

Status: accepted

Back to the list of papers