EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
2412. Enhancing the performance of carsharing systems by optimizing multi-attribute pricing plans
Invited abstract in session MB-59: Revenue Management in Sharing/Platform Economy, stream Pricing and Revenue Management.
Monday, 10:30-12:00Room: S08 (building: 101)
Authors (first author is the speaker)
| 1. | Masoud Golalikhani
|
| INESC TEC, Faculty of Engineering, University of Porto | |
| 2. | Beatriz Brito Oliveira
|
| INESC TEC, Faculty of Engineering, University of Porto | |
| 3. | Gonçalo Correia
|
| Transport & Planning, TUDelft | |
| 4. | José Fernando Oliveira
|
| INESC TEC, Faculty of Engineering, University of Porto | |
| 5. | Maria Antónia Carravilla
|
| INESC TEC, Faculty of Engineering, University of Porto |
Abstract
Carsharing operators commonly offer several pricing plans tailored for different customer profiles, constructed from various attributes such as registration, travel distance, and travel time fees. Each plan may have different rates across high/low demand periods and areas to increase fleet utilization and maintain a balanced supply-demand system. However, in most existing works, the heterogeneity of users is not considered, and plans consist of only a single attribute (i.e., travel time fee). In this study, we formulate a mixed-integer linear programming model with the objective of profit maximization to optimize multi-attribute plans with time- and location-dependent rates, considering customer preferences and travel behavior. A discrete choice model is incorporated to estimate customer response to the value of plan attributes. We utilize real-world data from the Brooklyn taxi trip dataset to validate results and provide managerial insights. Results demonstrate the effectiveness of this approach in improving system profitability, particularly through time- and location-dependent rates. The model can find the optimal or near-optimal solutions for real-sized problems within a reasonable time limit. Moreover, we developed an approximate algorithm to efficiently produce high-quality results for large-scale instances within a shorter timeframe. Finally, we delivered valuable managerial insights regarding customer segmentation, pricing plan complexity, and parameter sensitivity.
Keywords
- Revenue Management and Pricing
- Transportation
- Programming, Mixed-Integer
Status: accepted
Back to the list of papers