EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

4288. A Stochastic Shared Car Relocation Problem to Optimize Long Term Strategic Inventory Decisions

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. Mehmet Serdar Erdogan

Abstract


In one-way flexible car-sharing systems, customers can drop-off the rented car at any location inside the operation area of the company and price of the rental is proportional to distance or duration of the rental. Rules of this system result in a supply and demand imbalance since supply of each zone changes dynamically. High-demand zones tend to have lower supply, and low-demand zones tend to have higher supply. Therefore, car-sharing companies relocate the cars between zones to match the supply and demand. This relocation process is costly and requires optimization for a profitable business. In this study, we formulate a two-stage stochastic programming model to optimize the relocation decisions in a dynamic one-way flexible car-sharing system. The model decides inventory levels of each zone at the beginning of each time period, number of cars to be relocated between each pair of zones, and unsatisfied requests at each zone to minimize total cost of relocation and lost sales.This model is very effective for small and average size instances. To solve the large instances, we developed an approximate clustering method where zones are clustered into multiple clusters. This approximate clustering method can solve instances up to 40 zones and 36 time periods. To enhance the implementation, we have also proposed an extended model where the staff movement was taken into account. Results show that inventory levels at each zone tend to reach a steady-state.

Keywords

Status: accepted

Back to the list of papers