Link to the University of Pittsburgh Homepage
Link to the University Library System Homepage Link to the Contact Us Form

Models for excess demand in urban environments

Liu, Xin (2023) Models for excess demand in urban environments. Doctoral Dissertation, University of Pittsburgh. (Unpublished)

[img]
Preview
PDF
Updated Version

Download (12MB) | Preview

Abstract

In urban lives, citizens are motivated to visit business venues by personal needs and venue attractiveness. This creates the demand from citizens on urban businesses. As citizens move around the city to visit multiple business venues, they rely on the urban transportation systems. This creates the demand from citizens on transportation systems. To provide decent service, business venues and transportation systems are designed to satisfy a specific demand level per the operator's expectation. However, the actual demand can exceed the operator's expected demand level due to external factors (e.g., peak hour, weather, special venues nearby). The portion of the demand exceeding the operator's expected demand level is identified as the excess demand. Generally, existing works did not consider excess demand since such demand can easily be unobserved and ignored; this leads to biased analysis and forecasting for the actual demand.

In this thesis, firstly, we use the real-world data to uncover the existence of excess demand. Next, we estimate the excess demand for the urban business. Particularly, we propose our approach, which is based on simulations and complementarity, to estimate the excess demand for urban business entities. For each urban business venue, we estimate every source of its excess demand. For urban areas, we reveal the excess demand patterns among different periods in a day, and find that the excess demand can be explained by the venue diversity, venue density, number of venues and inter-area distance of urban areas. We fetch the embeddings of urban areas via a graph neural network and reveal the inter-area relationship in the latent space. Then, we estimate the excess demand of the urban transportation systems. Particularly, we propose our approach to estimate the excess demand in an urban bike sharing system. To predict the net total demand (which includes the observed and excess demand), we build a Skellam regression model, which shows advantages over other alternative models, both in terms of predictive performance and interpretability. Moreover, our Skellam regression model, as a generalized linear model, allows us to get a better estimation of the uncertainty of our prediction. The estimated excess demand provides insights for business owners, transportation operators and urban planners to satisfy more demand, which increases the revenue for business and creates more convenience for citizens.


Share

Citation/Export:
Social Networking:
Share |

Details

Item Type: University of Pittsburgh ETD
Status: Unpublished
Creators/Authors:
CreatorsEmailPitt UsernameORCID
Liu, Xinxil178@pitt.eduxil1780000-0001-8389-5310
ETD Committee:
TitleMemberEmail AddressPitt UsernameORCID
Committee ChairPelechrinis, Konstantinoskpele@pitt.edukpele
Committee MemberKrishnamurthy, Prashantprashk@pitt.eduprashk
Committee MemberKarimi, Hassanhkarimi@pitt.eduhkarimi
Committee MemberLabrinidis, Alexandroslabrinid@cs.pitt.edu
Date: 10 January 2023
Date Type: Publication
Defense Date: 11 November 2022
Approval Date: 10 January 2023
Submission Date: 30 November 2022
Access Restriction: No restriction; Release the ETD for access worldwide immediately.
Number of Pages: 124
Institution: University of Pittsburgh
Schools and Programs: School of Computing and Information > Telecommunications
Degree: PhD - Doctor of Philosophy
Thesis Type: Doctoral Dissertation
Refereed: Yes
Uncontrolled Keywords: Excess Demand, Urban Business, Urban Transportation
Date Deposited: 10 Jan 2023 16:16
Last Modified: 10 Jan 2023 16:16
URI: http://d-scholarship.pitt.edu/id/eprint/43911

Metrics

Monthly Views for the past 3 years

Plum Analytics


Actions (login required)

View Item View Item