University of California Institute of Transportation Studies

This research makes explicit and tests an implicit assumption in policies promoting public investment in plug-in electric vehicle (PEV) charging infrastructure: even people who are not already interested in PEVs see public PEV charging. Data from a survey representing all car-owning households in California are combined with per capita counts of public PEV charging locations and PEV registrations to estimate a structural equation model for two central variables: the extent to which participants have already considered acquiring a battery electric vehicle (BEV) or plug-in hybrid electric vehicle (PHEV), and whether and how many places people see PEV charging. The model controls for socio-economic and demographic measures as well as participants’ awareness, knowledge, and assessments of PEVs. The model also controls for the known spatial correlation between PEV registrations and public PEV charging locations. The conclusion is there is no evidence of a relationship between public charging location density and participants reporting they see PEV charging locations. Nor is there a relationship between public charging location density and PEV purchase consideration. The evidence indicates there is little reason to assume building more public PEV charging means more people will see that charging or that more people will consider purchasing a PEV. Rather, awareness, knowledge, and positive assessments of PEVs allow people to see PEV charging in their local environment. In short, interest in PEVs is a prerequisite to people seeing PEV charging. Concomitant investments to increase awareness of PEVs and engagement in a transition to them as well as in PEV charging infrastructure may be a more effective way to grow the PEV market.

Congestion pricing (CP) is widely considered to have significant potential for effectively reducing vehicle miles traveled, reducing emissions, and providing a reliable revenue source for transportation investments. This study evaluated cities interested in CP—five in the U.S. (Boston, Los Angeles, New York, San Francisco, Seattle) and two in other countries (Vancouver, Canada, and Auckland, New Zealand). This study examines the following features of a CP system for each of these cities: 1) duration of CP investigations, 2) equity mitigations, 3) range of alternatives considered, 4) public engagement, and 5) importance of emissions reductions. Timelines are impossible to predict with certainty, but New York and Auckland appear closest to implementation. Vancouver, San Francisco, and Seattle are well into the process; and Boston and Los Angeles are early in the process. Other key findings include that most of the cities start considering a range of options before narrowing down to comparing more detailed CP systems. Vancouver and San Francisco have made public engagement a cornerstone of their plan development, using polls and workshops to finetune the details of their CP proposals. In contrast, Auckland, while still engaging with stakeholders and experts for guidance, has mainly focused on how to ensure public support and understanding of the proposals they recommend. In terms of equity, discounts are a common and primary strategy proposed among the cities, but some also develop a more comprehensive set of equity policies to accompany a CP system.

While micromobility services (e.g., bikeshare, e-bike share, e-scooter share) hold great potential for providing clean travel, estimating the effects of those services on vehicle miles traveled and reducing greenhouse gases is challenging. To address some of the challenges, this study examined survey, micromobility, and transit data collected from 2017 to 2021 in approximately 20 U.S. cities. Micromobility fleet utilization ranged widely from 0.7 to 12 trips per vehicle per day, and the average trip distance was 0.8 to 3.6 miles. The median (range) rates at which micromobility trips substituted for other modes were 41% (16–71%) for car trips, 36% (5–48%) for walking, and 8% (2–35%) for transit, 5% (2–42%) for no trip. In most cities, the mean actual trip distance was approximately 1.5 to 2 times longer than the mean distance of a line connecting origin to destination. There was a weak and unclear connection between micromobility use and transit use that requires further study to more clearly delineate, but micromobility use had a stronger positive relationship to nearby rail use than to nearby bus use in cities with rail and bus service. The COVID-19 pandemic led to more moderate declines in docked than in dockless bike-share systems. Metrics that would enable better assessment of the impacts of micromobility are vehicle miles traveled and emissions of micromobility fleets and their service vehicles, and miles and percentage of micromobility trips that connect to transit or substitute for car trips.

California has set a goal of reaching 100% zero emission vehicle (ZEV) sales by 2035. Most ZEV sales to date have been battery electric vehicles (BEVs) or plug-in hybrid electric vehicles (PHEVs), while fuel cell electric vehicles (FCEVs) make up only a small portion of ZEV sales. The market for FCEVs may be partially constrained because, unlike BEVs and PHEVs, they cannot use any existing infrastructure. This research investigates FCEV drivers use of hydrogen stations in California (of which there are 47 in operation) with the goal of informing the development of hydrogen infrastructure. Hydrogen station use was studied using results from a 2017 survey of 395 fuel cell electric vehicle (FCEV) owners and a 2018 survey of 328 FCEV owners. The results show FCEV drivers use on average 2.4 hydrogen stations. The average shortest distance FCEV owners would need to travel from home, work, or their commute to a hydrogen refueling station was 10 miles. Those whose most-used station was not the closest station available were more likely than those whose most-used station was the closest to use renewable hydrogen, suggesting that some drivers may prefer renewable hydrogen. Currently the percentage of California census block groups with one, two, and three hydrogen stations within 10 miles of households are 52.4%, 25.6%, and 22.5%; these census block groups are concentrated primarily in large metropolitan areas. Finally, 70% of FCEV owners said they would not have purchased the vehicle if their primary station had not been available, pointing the importance of station availability to FCEV adoption.

Passenger and heavy-duty vehicles make up 36% of California’s greenhouse gas (GHG) emissions. Reducing emissions from vehicular travel is therefore paramount for any path towards carbon neutrality. Efforts to reduce GHGs by encouraging mode shift or increasing vehicle efficiency are, and will continue to be, a critical part of decarbonizing the transportation sector. Emerging technologies are creating an opportunity to reduce GHGs. Human driving behaviors in congested traffic have been shown to create stop-and-go waves. When waves form, cars periodically slow down (sometimes to a stop) and then speed back up again; this repeated braking and accelerating leads to higher fuel consumption, and correspondingly increasingly GHG emissions. Flow smoothing, or the use of a specially designed adaptive cruise controllers to dissipate these waves, can reduce fuel consumption of all the cars on the road. By keeping all vehicles at a constant speed, flow smoothing can minimize system-wide GHG emissions. This report presents the results of flow-smoothing when used in simulation, discusses current work on implementing flow-smoothing in real world-highways, and presents policy discussions on how to support flow smoothing.

During the COVID-19 outbreak, serious concerns were raised over the risk of spreading the infection on public transportation systems. As the pandemic recedes it will be important to determine optimal timetable design to minimize the risk of new infections as systems resume full service. In this study, we developed an integrated optimization model for service line reopening plans and timetable design. Our model combines a space-time passenger network flow problem and compartmental epidemiological models for each vehicle and platform in the transit system. The algorithm can help policy makers to design schedules under COVID-19 more efficiently. The report develops an optimized timetable for the Bay Area Rapid Transit system. We found that if passengers choose other mode of transportation when closing part of the system or decreasing the frequency of service can prevent the spread of infections, otherwise, if passengers choose to use the closest open station, closings will lead to longer waiting times, higher passenger density and greater infection risk. We found that the goal of stopping the spread of infection could be achieved by minimizing the total delay when infections were similar in different districts across the service area. Where infection rates are different in different districts, minimizing the risk of exposure can be achieved by minimizing weighted travel time where higher weights are applied to areas where the infection rate is highest.

Transportation and other agencies and organizations are increasingly planning and building under- and over-crossing structures for wildlife to traverse busy highways. However, if wildlife do not use these structures due to noise, light, and other factors, then the structures may have a low benefit to cost ratio. Several criteria are key for their success— sufficient safety and/or conservation need, cost, location, and anticipated use by wildlife. There is limited information in wildlife-crossing guidance on how wildlife biologists should advise designers, engineers, and architects on the use of structural and vegetation elements that could reduce noise and light disturbances. To address this problem, this study used field measurements and modeling of light and noise from traffic to inform and test the designs of two wildlife overcrossings. Wildlife-responsive designs were developed and tested for two crossings being considered or planned by California Department of Transportation in California. For the planned crossing of US 101 near the city of Agoura Hills (the Wallis-Annenberg crossing), the three designs consisted of noise/glare barriers; noise/glare barriers + berm; and noise/glare barriers + multiple berms. For the potential crossing of Interstate 15 south of Temecula, one design used noise/glare barriers of 3 different heights and the other had no barriers. Key limitations and opportunities for each design approach were identified. Creating “dark and quiet paths” using a combination of berms and noise/glare barriers could decrease disturbance in the crossing structure approach zones and increase the wildlife-responsiveness of the designs.

This project aims to provide recommendations on the methodology and design specifications for the travel demand model to be built for the Link21 program in the Northern California megaregion. The Link21 program is a major rail investment program that will considerably improve and upgrade the passenger rail services in the Northern California megaregion, centered around the Transbay Corridor between Oakland and San Francisco in the San Francisco Bay Area. To support this effort, we reviewed the current and potential travel markets for the Link21 program, assessed the available travel demand models that could be used to support the modeling efforts for the Link21 program, and conducted interviews with experts from academic institutions, metropolitan planning organizations, state and federal agencies, and US DOE national labs. Considering the goals and objectives of the Link21 program, a list of 20 critical, important, and optional modeling features were identified, which should be considered for the Link21 program. We reviewed 11 existing travel demand models based on the evaluation of their modeling features, and present four proposed modeling approaches which could be considered to support the Link21 program. For each modeling approach, we summarize pros and cons in terms of fulfilling the requirements of the Link21 program. The four modeling approaches include: 1) building on the Metropolitan Transportation Commission (MTC) TM 2.1 regional travel demand model without a dedicated long-distance travel model component; 2) building on the MTC TM 2.1 regional travel demand model with a dedicated long-distance travel model component; 3) building on the San Francisco County Transportation Authority (SFCTA) regional travel demand model with or without a dedicated long-distance travel model component; and 4) building on the California High Speed Rail (CHSR) or the new statewide rail model that is currently under development. The study also discusses some sources of uncertainties that might affect future travel demand and the modeling practice in the Link21 regions. These include the impacts of the COVID-19 pandemic on work patterns and activity/travel choices, the introduction of shared mobility services, micromobility, the potential deployment of Mobility as a Service (MaaS) solutions, and the forthcoming deployment of connected and automated vehicles (CAVs). Given the complexity of the Link21 program and the requested 18-month timeline for developing a new travel demand model to support the program, we recommend that the model development for the Link21 program build on an existing modeling framework and adopt a modular system, which can be updated over time. An initial model release would become available in the proposed timeline of 18 months, while future updates and improvements in the model components could be added in future model updates. This process also would be well-suited to address eventual modeling issues that could arise with the initial model release, and it would benefit from the development and updates of other models in the Northern California megaregion that are being carried out in parallel.

Pilot programs have been implemented in cities across the U.S. to address the first- and last- mile problem with door-to-door shared microtransit, ridehailing companies, and shared-ride operators with dynamic pick-up locations. The City of Rancho Cordova and Lyft partnered to launch one such pilot in the form of a discount-based door-to-door (D2D) coupon program named “Free ＄5 to Ride”. The program offers ＄5 credits to Lyft riders who start or end their trips at one of four Sacramento Regional Transit District (SacRT) light rail stations. The program was designed to reduce rider dependence on personal vehicles and increase the overall convenience of transit use in the region. UC Davis researchers conducted an evaluation of the “Free ＄5 to Ride” program during its operational period of May 2019 through June 2021. Researchers developed a participant survey and used survey data along with participant trip data, ridership data for the SacRT light rail, and ridership data for the Rancho CordoVan shuttle service to characterize the outcomes of the pilot program. The evaluation shows that the coupon program was generally well-received. Participation levels increased dramatically by early 2020, and while trip activity dropped at the onset of the COVID-19 pandemic, program activity remained fairly constant through the end of the program. Researchers encountered survey sampling limitations due to ridehailing customer engagement policies, suggesting that future evaluations of similar programs would benefit from increased data access, or modified policies allowing operators to conduct more extensive outreach in support of these studies.