Abstract
Urban transportation systems face ongoing challenges, including noise, air pollution, congestion, and the inefficient use of space. Shared micromobility services (SMSs), including both dockless and station-based bikes and scooters, offer a potential solution to these issues. By providing a flexible mobility alternative, SMSs not only improve urban accessibility but also reduce dependency on short car trips, which can help alleviate traffic congestion and emissions. However, SMSs have received criticism for serving a narrow demographic of residents and visitors while underserving underrepresented individuals. This limitation reduces SMSs’ ability to support a more equitable transportation system.To ensure equitable access to SMS, understanding demand patterns across different communities is crucial. This includes predicting where services will be most needed and ensuring a minimum vehicle supply in underserved areas. However, creating a reliable and equitable service is not straightforward. Each city or region where an SMS exists has unique factors, like geography, demographics, and land use, which all impact SMS demand and access needs. Moreover, cities have differing equity policies that attempt to increase access to SMS in disadvantaged communities.In light of this background, this dissertation addresses the following research questions:What are the key determinants of SMS ridership—both docked and dockless—identified in the existing literature, and how consistent are these factors across different cities and regions with varying urban and socioeconomic characteristics?How do these determinants influence dockless SMS ridership patterns within different zones of a city, and how do these patterns vary between small- and large-sized cities with distinct urban features?What are the existing dockless SMS equity policies, are there systematic biases in the availability of SMS vehicles, and to what extent do equity policies address inequities in SMS accessibility for underserved and minority populations?Chapter 2 of this dissertation identifies the critical factors influencing SMS ridership through a meta-analysis of the existing studies on SMS demand forecast. The meta-analysis aggregates and synthesizes findings from 29 empirical studies and aims to provide: a) a clearer understanding of which SMS ridership predictors have been validated in the literature and should be prioritized for SMS ridership forecast and b) estimates of how these factors impact SMS usage. This meta-analysis serves as a foundation for Chapter 4 of this dissertation.In Chapter 3, the datasets and preprocessing methods that underpin the analyses in subsequent chapters are introduced. This chapter provides a comprehensive overview of the data sources, including Spin’s SMS ridership and operation data, sociodemographic attributes from the American Community Survey (ACS), land use and built environment metrics from the EPA’s Smart Location Database (SLD), transit infrastructure data from General Transit Feed Specifications (GTFS), and points-of-interest (POIs) from Google Places. By combining spatial data, equity policies, and performance metrics, this chapter lays the foundation for understanding how sociodemographic, built environment, and infrastructure variables shape SMS usage and access patterns.Chapter 4 of this dissertation employs spatial regression models, specifically spatial lag models, to analyze dockless SMS ridership across three U.S. cities with diverse urban characteristics: Washington D.C., San Francisco, and Fort Collins. This chapter examines how sociodemographic, built environment, transit supply, and infrastructure factors influence SMS demand, providing city-specific insights into the determinants of ridership. The methodological approach allows for the exploration of spatial dependencies and spillover effects. The findings also reveal significant variability in how these factors impact SMS ridership across cities. By comparing the elasticity findings to meta-analyses of docked SMS systems, this chapter highlights both universal and context-specific drivers of SMS demand. Core factors such as income, population density, employment density, and transit accessibility remain critical across system types. However, the observed variations across cities underscore the necessity of localized SMS strategies tailored to unique urban dynamics.In Chapter 5, this dissertation investigates the distributional equity of dockless SMS in Washington D.C., focusing on the relationship between reliable access to SMS vehicles and socio-demographic, land use, and built environment characteristics. Two proposed metrics for reliable SMS access are developed, combining acceptable walking distances with the number of nearby available scooters. Using data from SMS provider Spin, spatial analyses are conducted to quantify disparities in SMS vehicle access across neighborhoods. The results reveal that areas with higher poverty ratios are consistently underserved, while neighborhoods with higher non-white populations are oversupplied, highlighting inequities in vehicle accessibility. Additionally, areas with an abundance of activity locations (parks, museums, restaurants, and hotels) often face undersupply.