Resource allocation and pricing under competition in shared mobility markets

Jiang, Zhoutong

Permalink

https://hdl.handle.net/2142/113177 Copy

Description

Title

Resource allocation and pricing under competition in shared mobility markets

Author(s)

Jiang, Zhoutong

Issue Date

2021-07-13

Director of Research (if dissertation) or Advisor (if thesis)

Ouyang, Yanfeng

Doctoral Committee Chair(s)

Ouyang, Yanfeng

Committee Member(s)

• Lee, Bumsoo
• Meidani, Hadi
• Lehe, Lewis

Department of Study

Civil & Environmental Eng

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Shared mobility
• Ridesharing
• Bikesharing
• Competition
• Pricing
• Resource allocation
• Approximate dynamic programming
• Stochastic programming

Abstract

Shared mobility services, such as those involving shared rides and shared cars/bikes, are becoming important components of urban transportation systems. There are various advantages of shared mobility, including: (i) significant societal benefit to our environment; for example, bikesharing is considered as a promising way to reduce traffic congestion, harmful gas emissions, and fuel consumption (Qiu and He, 2018); (ii) economic benefits to users, by reducing their travel costs especially those related to owning and maintaining a vehicle; (iii) increased accessibility of users, e.g., by bridging the spatiotemporal gaps in existing transportation service networks (Shaheen et al., 2016); and (iv) creation of new job opportunities; for example, in 2014, the Uber platform created 20,000 new jobs each month (Eadicicco, 2014). Nevertheless, fierce competition and lack of strategic planning have led to repeated failures of shared mobility services in many cities. For example, oversupply and poor management of bikes – partly due to competition among bikesharing companies – have proven to be counter productive, causing huge wastes of resources and significant societal disbenefits in some countries. The service providers face lots of challenges in their daily operations. For example, in a ridesharing market, significant imbalance between spatiotemporal distributions of vehicle supply and travel demand mandates companies to allocate their resources (e.g., vehicles and drivers) and design pricing strategies optimally to incentivize demand and maximize their profits. It is even more challenging for the service providers to make optimal operation decisions when there are multiple companies in a shared mobility market. For example, in a bikesharing market, where the market share depends on how the competing companies deploy their bikes over time and space as well as how they make pricing decisions, each individual company needs to maximize its profit through optimal pricing, investment, and allocation/rebalancing strategies, in response to not only time-varying demand but also actions of the fellow competitors. All the above challenges highlight the urgent needs for a systematic modeling framework to analyze the optimal investment, management, and pricing strategies for these companies under competition. This Ph.D. dissertation aims at investigating several important topics in competitive shared mobility markets (both ridesharing and vehicle sharing), including: (i) dynamic pricing and resource allocation in an on-demand ridesharing market; (ii) pricing and matching in a two-sided ridesharing market under competition; (iii) optimal investment and management of dockless shared bikes in a competitive market; and (iv) pricing and resource allocation in a docked bikesharing market under competition. First, we propose a multi-period game-theoretic model that addresses dynamic pricing and idling vehicle dispatching problems in a one-sided ridesharing market (i.e., with fully compliant drivers/vehicles). A dynamic mathematical program with equilibrium constraints (MPEC) is formulated to capture the interdependent decision-making processes of the mobility service provider (e.g., regarding vehicle allocation) and travelers (e.g., regarding ridesharing and travel path options). An algorithm based on approximate dynamic programming (ADP), with customized subroutines for solving the MPEC, is developed to solve the overall problem. It is shown with numerical experiments that the proposed dynamic pricing and vehicle dispatching strategy can help ridesharing service providers achieve better system performance (as compared with myopic policies) while facing spatial and temporal variations in ridesharing demand. Then we study the competition between two companies in a two-sided ridesharing market. The two companies share and compete for drivers and riders, and each company optimizes its pricing and driver-rider matching strategies to maximize its profit. The competition is modeled as a generalized Nash equilibrium problem (GNEP). We consider the independent decision-making process of drivers and riders, and show that the game is a potential game which can be solved by systematic approaches. We then investigate the impact of competition and draw managerial insights through a series of hypothetical numerical experiments. Next, we shift our attention from ridersharing to bikesharing. We develop a game-theoretical framework to model the competition between two bikesharing companies in a dockless bikesharing market. A two-stage multi-period stochastic program is developed to model the decision process of each company regarding the number and spatiotemporal distribution of bikes in a city. The effects of demand elasticity and uncertainty are also discussed. We then show the existence of Nash equilibrium and analytical insights into the solution for several special cases. For general cases, an iterative algorithm is proposed to solve the Nash equilibrium. Numerical experiments are conducted to demonstrate the applicability of the proposed model and to draw insights into the impacts of market competition. Finally, we consider the competition between two companies in a docked bikesharing market, where the additional decisions on dock-station installation and pricing decisions are also addressed. We model the competition as a GNEP, in which the users’ behaviors are modeled using a set of logical constraints. We then develop reformulation approaches to convert the nonlinear model into a linear one and show how the equilibrium of the GNEP can be obtained. We demonstrate the optimal decisions for the companies, and investigate the market property under competition through hypothetical and real-world case studies.

Graduation Semester

2021-08

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/113177

Copyright and License Information

Copyright 2021 Zhoutong Jiang

Owning Collections