EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

3966. A robust satisficing multi-objective optimization approach for bike-sharing systems with heterogeneous user types

Invited abstract in session WC-34: Transportation and Logistics under Uncertainty, stream Stochastic, Robust and Distributionally Robust Optimization.

Wednesday, 12:30-14:00
Room: 43 (building: 303A)

Authors (first author is the speaker)

1. Qingxin Chen
College of Management and Economics, Tianjin University
2. Shoufeng Ma
College of Management and Economics, Tianjin University
3. Ning Zhu
Tianjin University

Abstract

Recently bike-sharing systems begin to employ diverse pricing and subscription policies to cater to the commuting needs of heterogeneous users. Frequent riders choose to purchase subscriptions to reduce their commuting costs if the system is easily accessible. To maintain a high market share, operators should improve their service level to retain a large number of subscribers. In addition, there is also a small proportion of general customers who use the system occasionally. These customers contribute to the operational profits, and providers need to fulfill as much customer demand as possible to ensure high profitability. Thus, maintaining the service level and obtaining profits are both important operational objectives in bike-sharing systems. However, jointly optimizing these objectives for the heterogeneous users in bike-sharing systems is challenging, especially when demand is highly uncertain. To address these issues, this study presents a robust satisficing optimization model that jointly optimizes the service level and profits in bike-sharing systems under demand uncertainty. To avoid over-conservation of solutions, side information like weather and weekend is integrated into the model to describe the relationship between demand and these exogenous factors. Extensive numerical experiments show that our model achieves higher robustness, lower violation probability and degree, and higher average-case out-of-sample performance than other benchmarks.

Keywords

Status: accepted

Back to the list of papers