UCLA

Urbanization accelerates greenhouse gas emissions and traffic congestion, endangering the environmental sustainability of cities and causing significant economic losses. In this thesis, we aim to tackle practical problems faced by public sector by leveraging tools including economic modeling, optimization, and data analytics to achieve sustainable urban mobility.

In Chapter 2, we focus on an incentive mechanism design problem to improve public transportation adoption. Due to a prolonged decline in public transit ridership over the last decade, transit agencies across the United States are in financial crisis. To entice commuters to travel by public transit instead of driving personal vehicles, municipal governments must address the “last mile” problem by providing convenient and affordable transportation between a commuter’s home and a transit station. This challenge raises an important question: Is there a cost-effective subsidy program that can improve public transit adoption by solving the last-mile problem? To address the question, we present and analyze two incentive mechanisms applied in practice for increasing commuter adoption of public transit. In a direct mechanism, the government provides a subsidy to commuters who adopt a “mixed mode”, which involves taking public transit and hailing rides to/from a transit station. The government funds the subsidy by imposing congestion fees on personal vehicles entering the city center. In an indirect mechanism, instead of levying congestion fees, the government secures funding for the subsidy from the private sector. These two mechanisms are especially relevant because several jurisdictions in the U.S. have begun piloting incentive programs in which commuters receive subsidies for ride-hailing trips that begin or end at a transit station. We then examine the implications of both mechanisms on five self-interested stakeholders (commuters, public transit agency, ride-hailing platform, municipal government, and local private enterprises), and provide conditions where either mechanism is superior with respect to total cost and commuter welfare. Our findings offer cost-effective prescriptions to municipal governments seeking to improve urban mobility and public transit ridership.

Chapter 3 of the thesis delves into improving cycling ridership through bike lane network planning. Sustainable urban infrastructure plays a pivotal role in the creation of livable cities. To meet the growing demand for cycling and reduce emissions, municipal governments worldwide have made significant investments in the expansion of bike lane networks. However, re-allocating road capacity from vehicles to cycling can often prove controversial due to the risk of exacerbating traffic congestion. This chapter presents a method for bike lane network planning that accounts for ridership and congestion effects. We first develop a game theoretic model that captures the traffic equilibrium and consumers’ transportation mode choice. We then present an estimator for recovering unknown parameters of a traffic equilibrium model from features of a road network and observed vehicle flows, which we show asymptotically recovers ground-truth parameters as the network grows large. We subsequently present a prescriptive model that recommends paths in a road network for bike lane construction while endogenizing cycling demand, driver route choice, and driving travel times. Our framework allows for a thorough assessment of bike lane network expansion on both cycling ridership and traffic congestion, which offers valuable insights and solutions for social planners seeking to promote urban mobility via urban infrastructure planning.

In Chapter 4, we build on the theoretical foundation of Chapter 3 and conduct a comprehensive empirical study on the City of Chicago. We bring together data on the road and bike lane networks, vehicle flows, travel mode choices, bike share trips, driving and cycling routes, and taxi trips to estimate the impact of expanding Chicago’s bike lane network. We estimate that adding 25 miles of bike lanes as prescribed by our model can lift ridership from 3.9% to 6.9%, with at most an 8% increase in driving times. We also find that three intuitive heuristics for bike lane planning can lead to lower ridership and worse congestion outcomes, which highlights the value of a holistic and data-driven approach to urban infrastructure planning.