website

Freeway Revolts and Racially Exclusive Participatory Planning: A History of Organized Opposition to Freeway Construction in Pacoima

Publication Date

May 8, 2025

Author(s)

Anastasia Loukaitou-Sideris, Susan Handy, Paul Ong, Jesus M. Barajas, Jacob Wasserman, Chhandara Pech, Juan Carlos Garcia Sanchez, Andres F Ramirez, Aakansha Jain, Emmanuel Proussaloglou, Andrea Nguyen, Katherine Turner, Abigail Fitzgibbon, Francois Kaeppelin, Felipe Ramirez, Marc Arenas

Abstract

Located in northeast Los Angeles, Pacoima is one of the oldest neighborhoods in the San Fernando Valley. Today, it is also one of the most polluted. Within the neighborhood’s 4.3 square miles, Pacoima contains three freeways, a railroad line, a small airport, and more than 300 industrial facilities. Before the construction of the freeways beginning in the 1960s, however, the neighborhood looked very different. Once known as America’s “first Black suburb,” Pacoima had a rich history of activism; however, archival records of the time offer little to no evidence of pushback from Black residents to the State Division of Highways (the precursor to Caltrans), as the Simi Freeway/State Route 118 bulldozed through their community in the 1960s. This storymap looks into neighboring opposition that led to the chosen route cutting through Pacoima’s vibrant and diverse community.

website

From "Across the Tracks" to "Across the Freeway": A History of the Racialization and Exclusion behind the I-10 Freeway in Colton

Publication Date

May 30, 2025

Author(s)

Paul Ong, Chhandara Pech, Jacob Wasserman, Andres F Ramirez, Anastasia Loukaitou-Sideris, Leila Ullmann, Megan Riley

Abstract

Travel 62 miles east of Los Angeles and you will find an epicenter of global mobility. Colton, a small 16-square-mile city in San Bernardino County, has been a principal rail and logistics hub since its establishment in the late 19th century. More than 110 cargo trains pass through the city daily, and it sits amidst the many warehouses of the Inland Empire region that store imports before their distribution locally and nationwide. But for Colton’s communities of color, especially its Latino residents, these transportation systems have long been a double-edged sword — offering employment opportunities while also enforcing stark racial and spatial divides. Railroad tracks historically split the city in two: white, affluent North Colton and working-class, Latino South Colton. In the 1940s, when state officials began planning the San Bernardino Freeway/Interstate 10 route, the proposed alignment threatened to deepen these divisions and further marginalize South Colton. This storymap looks into the nearly decade of heated debate that ultimately led to a route that minimized direct displacement and racial impact — a rare outcome in freeway planning of the era, driven largely by cost considerations.

preprint journal article

MobilityGPT: Enhanced Human Mobility Modeling with a GPT model

Abstract

Generative models have shown promising results in capturing human mobility characteristics and generating synthetic trajectories. However, it remains challenging to ensure that the generated geospatial mobility data is semantically realistic, including consistent location sequences, and reflects real-world characteristics, such as constraining on geospatial limits. This project reformats human mobility modeling as an autoregressive generation task to address these issues, leveraging the Generative Pre-trained Transformer (GPT) architecture. To ensure its controllable generation to alleviate the above challenges, the study proposes a geospatially-aware generative model, MobilityGPT. A gravity-based sampling method is proposed to train a transformer for semantic sequence similarity. The training process would be constrained via a road connectivity matrix that provides the connectivity of sequences in trajectory generation, thereby keeping generated trajectories in geospatial limits. Lastly, The study proposed to construct a preference dataset for fine-tuning MobilityGPT via Reinforcement Learning from Trajectory Feedback (RLTF) mechanism, which minimizes the travel distance between training and the synthetically generated trajectories. Experiments on real-world datasets demonstrate MobilityGPT’s superior performance over state-of-the-art methods in generating high-quality mobility trajectories that are closest to real data in terms of origin-destination similarity, trip length, travel radius, link, and gravity distributions.

policy brief

Examining Both Trip Level Mode Replacements and Daily Activity Patterns of Users is Required to Understand the Sustainability Potential of Micromobility

Abstract

Micromobility options such as electric bike-share and scooter-share services are a fundamental part of the existing shared mobility landscape. Research has shown that micromobility use can reduce car dependence. This is accomplished through trip-level mode replacement and adjustments in mode-use configurations in daily travel. Understanding the full potential of micromobility services as a car replacement can help cities better plan for the services to meet environmental sustainability goals. Researchers at the University of California, Davis collected GPS-based travel diary data from individual micromobility users from 48 cities in the US and examined their travel behavior and micromobility use patterns. They found that micromobility services can displace car use. To achieve environmental sustainability goals, cities must pursue options that will deliver benefits, such as micromobility services. This policy brief summarizes the findings from that research and provides policy implications.

research report

Heterogeneity of Plug-in Electric Vehicle Owners in Rural California

Publication Date

September 1, 2025

Author(s)

Anya Robinson, Theodora Konstantinou, Gil Tal

Abstract

Little is known about plug-in electric vehicle (PEV) ownership, charging behavior, and vehicle characteristics in rural California. As the state works toward its goal of carbon neutrality by 2045, understanding the current state of PEV adoption in rural areas is essential for identifying where targeted support may be needed to meet electrification objectives. Existing definitions of “rural” may also obscure important variation within these regions. This study proposes a passenger-vehicle-based classification of rural areas in the state using k-means clustering, incorporating data on land use, travel behavior, vehicle characteristics, and housing attributes. Five distinct clusters were identified, three of which – Rural Remote, Farm Rural, and Small Town – were classified as rural. Survey data from PEV and conventional vehicle (CV) owners were analyzed to compare sociodemographic characteristics, vehicle attributes, and charging access and behavior. Across all clusters, PEVs were newer and generally smaller than CVs. The Rural Remote cluster exhibited the highest rural PEV adoption rates (1.4% BEVs, 1.0% PHEVs), along with higher household income, education, and Level 2 home charging prevalence. Farm Rural and Small Town clusters had lower adoption rates and relied more heavily on Level 1 charging, despite comparable at-home charging frequency. Public charging access per capita was lowest in Rural Remote areas and highest in Small Town clusters across rural areas. These findings indicate that rural California is heterogeneous with respect to PEV ownership and future adoption potential. Policies that account for demographic, infrastructural, and travel behavior differences between rural subtypes may be more effective than uniform approaches in supporting adoption.

policy brief

Traffic Collisions Change How Victims Think About Safety

Publication Date

September 1, 2025

Author(s)

Md. Musfiqur Rahman Bhuiya, Jesus M. Barajas, Prashanth Venkataram

Abstract

Traffic safety remains a pressing concern in California. Over the past five years, the state has averaged more than 3,751 reported traffic fatalities annually, with likely more unreported. While policies and research often focus on crash prevention and severity reduction, less is known about how collisions affect individuals’ travel behavior and perceptions of road safety. To better understand these effects, the research team conducted interviews and focus groups with people who had direct or indirect experience with traffic collisions and near misses. The researchers also spoke with professionals who support collision victims, such as physicians, therapists, faith leaders, and advocacy groups representatives. Discussions focused on perceptions of road safety, transportation mode choices, and travel behavior of someone involved in a traffic collision or near miss before and after the incident.

policy brief

High-Occupancy Toll Lanes Can Improve Driving for Everyone if Dynamically Priced

Abstract

High occupancy vehicle (HOV) lanes allow carpool vehicles to bypass congestion and save travel time. However, HOV lanes are often underutilized, leading to a waste of road capacity or the loss of travel time advantage over general purpose lanes. To address this issue, high-occupancy toll (HOT) lanes have gained popularity. HOT lanes, when properly priced, preserve the advantage of HOV lanes, while allowing single-occupant vehicles to pay for access, making use of spare HOT lane capacity. To inform better pricing and utilization of HOT lanes, this study analyzed traffic data from both HOT and general purpose lanes to identify patterns in how toll rates impact traffic flow.

conference paper

Impact of Flight Trajectory Design on Performance and Noise for AAM Aircraft

Publication Date

June 1, 2024

Author(s)

Victoria Pellerito, Nathan Yeung, Jacqueline Huynh

Abstract

Advanced Air Mobility (AAM) is an evolving field of research seeking to transform sustainable air transportation in urban and sub-urban environments amid increasing urbanization and traffic congestion. The evolution of AAM requires efficient management of congested airspace and the accommodation of diverse vehicles with distinct performance capabilities. A broad range of AAM aircraft are in development which will have different community noise footprints and energy use depending on the details of the departure and arrival flight trajectories which must be understood for airspace integration. This work presents a framework for analyzing AAM trajectory design, focusing on key performance characteristics including community noise impact, energy consumption, and flight duration. The framework can be applied to diverse AAM vehicle types, as demonstrated in this work on a Blown-Flap Short Takeoff and Landing vehicle, a Tilt-Rotor Vertical Takeoff and Landing vehicle, and a Lift Plus Cruise Vertical Takeoff and Landing vehicle. Results of comparing various takeoff procedures for each vehicle show trade-offs between community noise, energy consumption, and flight duration, highlighting the importance of strategic trajectory design.

research report

Vehicle Purchasing Behavior, Expenditure, and Potential Barriers to Uptake of Battery Electric Vehicles in Underserved Communities

Abstract

Plug-in electric vehicles (PEVs), including both battery-electric vehicles (BEVs) and plug-in hybrid electric vehicles (PHEVs) are crucial forreducing emissions and meeting sustainability targets, yet their adoption has been limited primarily to higher-income and new car buyers, leaving most low-income households without access. To help inform policies that will accelerate access to used BEVs in particular, this study explored car buying behavior, costs, and usage within and between groups defined by vehicle condition (new vs. used), fuel type(battery-electric vehicles [BEVs] vs. internal combustion engine vehicles [ICEVs], and income level. BEV owning, new-car buying, andhaving higher incomewere each associated with one another. On average the proportion of total income spent on vehicle-relatedexpenses is at least six times higher for householdswith incomes less than $75,000 than householdswith incomes of $250,000 or more. While BEVs offer savings in maintenance and fuel cost compared to ICEVs, the initial price for both new and used BEVs may need to be subsidized to alleviate cost burden for lower-income households. Used car buyers, ICEV owners, and lower-income householdspredominantly do not purchase or maintain their vehicles at automaker dealerships and tend to buy older vehicles withmore mileagethan would be covered by BEV warranties. These findings have implications for the current structure of financial incentives being limited to automaker dealerships. Other possible barriers to BEV uptake for lower-income households and used car buyers, include reliability concerns and limited home charging access. BEV adoption across all income groups could be increased by broadening eligibility for incentives, enhancing battery warranties, offering battery replacement rebates, and expanding home charging infrastructure.

preprint journal article

Provably Safe and Human-Like Car-Following Behaviors: Part 1. Analysis of Phases and Dynamics in Standard Models

Abstract

Trajectory planning is essential for ensuring safe driving in the face of uncertainties related to communication, sensing, and dynamic factors such as weather, road conditions, policies, and other road users. Existing car-following models often lack rigorous safety proofs and the ability to replicate human-like driving behaviors consistently. This article applies multi-phase dynamical systems analysis to well-known car-following models to highlight the characteristics and limitations of existing approaches. It begins by formulating fundamental principles for safe and human-like car-following behaviors, which include zeroth-order principles for comfort and minimum jam spacings, first-order principles for speeds and time gaps, and second-order principles for comfort acceleration/deceleration bounds as well as braking profiles. From a set of these zeroth- and first-order principles, it derives Newell’s simplified car-following model. Subsequently, the study analyzes phases within the speed-spacing plane for the stationary lead-vehicle problem in Newell’s model and its extensions, which incorporate both bounded acceleration and deceleration. It then analyzes the performance of the Intelligent Driver Model and the Gipps model. Through this analysis, the study highlights the limitations of these models with respect to some of the aforementioned principles. Numerical simulations and empirical observations validate the theoretical insights. Finally, it discuss future research directions to further integrate safety, human-like behaviors, and vehicular automation in car-following models, which are addressed in Part 2 of this study, where it develops a novel multi-phase projection-based car-following model that addresses the limitations identified there.