Managing multiple services on a ride-hailing platform

In this thesis, we present our research on several operations management problems faced by a ride-hailing platform which provides multiple ride services. Firstly, we focus on carpooling services, which is the major substitution for regular non-pooling ride services. We empirically estimate the total welfare created by carpooling services during rush hour using a dataset from a leading platform in China. In particular, we model riders' choices among the carpooling, the regular and outside options, and derive a real-time carpooling pricing scheme that can maximize the total welfare. The results show that the carpooling service significantly increases riders' consumer surplus and can provide an effective way to mitigate supply constraint on a ride-hailing platform. Secondly, during a demand shock, when the supply becomes an exogenous limit, we consider an infinitesimal market where we describe the platform's supply allocation strategies and the quality trade-off among multiple services. We model price-and-time-sensitive riders on the platform using the choice model verified in the empirical study. The optimal supply allocation problem at market equilibrium is formulated and transformed into the market-share space, which turns out to be convex. Our sensitivity analysis then shows that while optimizing the profit, services with a greater profit margin or higher efficiency on driver usage shall have less waiting time. We further solve the joint pricing and supply allocation problem analytically. The result shows that it is optimal to serve the riders according to a descending order of the efficiency of their selected services, whereas the prices charged to more efficient services are higher in terms of cost rate. In the third part, we work out efficient algorithms for solving the optimization problem in the market-share space and constructing compatible allocation functions. In particular, the special structure of the feasible space for expected waiting time is investigated, and an O(N) time-and-space-complexity decomposition algorithm is provided. Numerical studies are also conducted to demonstrate the importance of considering the quality trade-off/supply allocation among multiple services. Finally, we generalize the single-type-driver model to the multiple-type-driver setting. We then analyze the supply allocation and the joint problem (with pricing), and present managerial insights.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Industrial Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Wang, Shuanglong

Contributors dc:contributor

• Chen, Xin
• Stolyar, Alexander
• Ouyang, Yanfeng
• Sun, Ruoyu

• Ride-Hailing Platform, Substitutable Services, Choice Model, Resource Allocation, Pricing

Statement dc:rights

• Copyright 2019 Shuanglong Wang

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/106449

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F106449