In 2017, California passed Senate Bill 1 (SB1) to bolster transportation infrastructure funding. Using data primarily from the California Department of Transportation (Caltrans)’s official SB1 progress reports, we analyze the severity of cost overruns, delays, and cancellations across SB1 Transportation Projects. Although events such as the COVID-19 pandemic likely caused some of these negative outcomes, our statistical models show consistent patterns of overruns associated with fiscal periods, programs, and geographic locations. Our results indicate that the common 20% contingency is generally insufficient, indicating the need for better risk estimation in project planning. We also suggest amplifying data transparency on project performance and re-evaluating project selection criteria to avoid rewarding underestimation of project costs and duration and penalizing accurate estimation.
While transportation infrastructure and efficiency should inform where to build more housing, little is known about how housing allocation and development processes can be coordinated more systematically with transportation. To date, transportation-housing coordination has often relied on the densification of areas near rail transit stations, putting heavy burdens on these locations and their residents. Much less attention has been paid to how densification can be achieved in a more equitable manner by encompassing other sites. This report directs attention to non-rail locations, specifically low vehicle miles traveled (VMT) areas and bus corridors, and examines the challenges that can arise in promoting densification more broadly. It shows that data uncertainties can make it challenging to identify low VMT locations and that prioritizing only low VMT locations for residential development may have limited effectiveness in expanding housing opportunities in high opportunity areas. The report further explores ways to achieve more inclusive densification of non-rail transit areas and highlights the importance of anti-displacement strategies.
Autonomous vehicles (AVs) heavily rely on machine learning-based perception models to accurately interpret their surroundings. However, these crucial perception components are vulnerable to a range of malicious attacks. Even though individual attacks can be highly successful, the actual security risks such attacks can pose to our daily life are unclear. Various factors, such as lack of stealthiness, cost-effectiveness, and ease of deployment, can deter potential attackers from employing certain attacks, thereby reducing the actual risk. This research report presents the first quantitative risk assessment for physical adversarial attacks on AVs. The specific focus is on attacks on AV’s perception components due to their highly critical function and representation in existing research. The report defines the daily-life risk as the likelihood that a given type of attack will be employed in real life and the authors develop a problem-specific risk scoring system and accompanying metrics. They perform an initial evaluation of the proposed risk assessment method for all the reported attacks on AVs from 2017 to 2023. They quantitatively rank the daily-life risks posed by each of eight different categories of attacks s and find three attacks with the highest risks: 2D printed images, 2D patches, and coated camouflage stickers, which deserve more focused attention for potential future mitigation strategy development and policy making.
Transportation pricing policies aim to manage vehicular demand for parking, dense urban areas, roadways, and highway lanes. Although pricing policies take various forms, most were designed in a world before the sharing economy and ride-sourcing companies. Hence, the efficacy of existing pricing policies in a world with shared mobility services requires further consideration. Moreover, future pricing policies designed to handle private vehicles and shared ride-sourcing vehicles must consider the behavior of both sets of travelers and vehicle fleets. This study develops a conceptual framework to support systems level analysis of pricing policies for a world with private and shared vehicle usage. It qualitatively analyzes the impact of shared vehicles on the effectiveness of various pricing policies, while also considering the role of vehicle-to-infrastructure technology. This conceptual framework will support future research that uses activity-based travel demand and dynamic network assignment models to evaluate congestion pricing policies in an era of shared mobility. Additionally, the study presents a detailed review of the literature related to transportation pricing together with a trend analysis on congestion pricing policies in Transportation Research Board annual meeting titles and abstracts.
The COVID-19 pandemic has generated renewed interest in how telecommuting can alter the workings of our cities and regions, but there is little guidance on how to align planning practice with the new reality. This report synthesizes the research on telecommuting and its consequences to help planners better understand what effects may occur from the proliferation of telecommuting and what lessons can be drawn from research findings. Emphasis is on the broad relevance of telecommuting to many domains of planning, including housing, land use, community development, and inclusive place-making, while attention is paid to changes in travel demand, vehicle miles traveled (VMT), and greenhouse gas emissions. The research suggests that telecommuting can occur in a variety of ways, and its impacts are largely dependent not only on the type/schedule of telecommuting but on the built environment, transit accessibility, and other amenities/opportunities the location provides. The varying impacts reported in the research can be seen as an encouragement for planners to actively create a better future rather than merely responding to the rise of telecommuting. Given the breadth of telecommuting’s impacts, systematic coordination across various planning domains will be increasingly important. This report also calls for collaboration across cities to guide the ongoing transformation induced by telecommuting not in a way that leads to more residential segregation but in a way that provides more sustainable and inclusive communities.
The upcoming Connected Vehicles (CV) technology shows great promise in effectively managing traffic congestion and enhancing mobility for users along transportation corridors. Data analysis powered by sensors in CVs allows us to implement optimized traffic management strategies optimizing the efficiency of transportation infrastructure resources. In this study, we introduce a novel Integrated Corridor Management (ICM) methodology, which integrates underutilized Park-And-Ride (PAR) facilities into the global optimization strategy. To achieve this, we use vehicle-to-infrastructure (V2I) communication protocols, namely basic safety messages (BSM) and traveler information messages (TIM) to help gather downstream traffic information and share park and ride advisories with upstream traffic, respectively. Next, we develop a model that assesses potential delays experienced by vehicles in the corridor. Based on this model, we employ a novel centralized deep reinforcement learning (DRL) solution to control the timing and content of these messages. The ultimate goal is to maximize throughput, minimize carbon emissions, and reduce travel time effectively. To evaluate our ICM strategy, we conduct simulations on a realistic model of Interstate 5 using the Veins simulation software. The DRL agent converges to a strategy that marginally improves throughput, travel speed, and freeway travel time, at the cost of a slightly higher carbon footprint.
Connected Vehicles (CV) technology offers significant potential for managing traffic congestion and improving mobility along transportation corridors. This report presents a novel approach using integrated corridor management (ICM) technology to divert CVs to underutilized park-and-ride facilities where drivers can park their vehicle and access public transportation. Using vehicle-to-infrastructure (V2I) communication protocols, the system collects data on downstream traffic and sends messages regarding available park-and-ride options to upstream traffic. A deep reinforcement learning (DRL) program controls the messaging, with the objective of maximizing traffic throughput and minimizing CO2 emissions and travel time. The ICM strategy is simulated on a realistic model of Interstate 5 using Veins simulation software. The results show marginal improvement in throughput, freeway travel time, and CO2 emissions, but increased travel delay for drivers choosing to divert to a park-and-ride facility to take public transportation for a portion of their travel.
The Federal Highway Administration (FHWA) vehicle classification scheme is designed to serve various transportation operational and planning needs. Many transportation agencies rely on Weigh-In-Motion and automatic vehicle classification sites to collect vehicle classification count data. However, these systems are not widely deployed due to high installation and operations costs. One cost-effective approach investigated by researchers has been the use of single inductive loop sensors as an alternative to obtain FHWA vehicle classification data. However, most models do not accurately classify under-represented classes, even though many of these minority classes pose disproportionally adverse impacts on pavement infrastructure and the environment. As a consequence, previous models have not been able to adequately classify under-represented classes, and the overall performance of the models are often masked by excellent classification accuracy of the majority classes, such as passenger vehicles and five-axle tractor trailers. This project developed a bootstrap aggregating (bagging) deep neural network (DNN) model on a truck-focused dataset obtained from Truck Activity Monitoring System (TAMS) sites, which leverage existing inductive loop sensor infrastructure coupled with deployed inductive loop signature technology, and already deployed statewide at over ninety locations across all Caltrans Districts. The proposed method significantly improved the model performance on truck-related classes, especially minority classes such as Classes 7 and 11 which were overlooked in previous research studies. Remarkably, the proposed model is also capable of distinguishing classes with overlapping axle configuration, which is generally a challenge for axle-based sensor systems.
Recent state legislation addresses California’s housing affordability crisis by encouraging new development in transitaccessible and/or jobs-rich areas. But policymakers lack key information about the effects of laws and plans on developers’ decisions about whether and where to build housing, and factors contributing to delays in receiving government development approvals in target areas. Drawing on a unique dataset detailing all residential projects of five units or more that were approved from 2014 through 2017 in selected California jurisdictions, this project analyzes how project attributes and transportation-related factors affected infill housing construction. We find that in cities with extensive transit infrastructure, new projects were generally located in parts of the city with high proximity to transit, but that proximity to rail stops or high frequency bus stops was not associated with extreme delays in project approval compared to all projects in general. The only factors related to extreme delay are the percentage of land within a half mile radius of dedicated single-family housing and whether a multiunit project required a rezoning or general plan amendment, the latter of which is associated with 326% increase in the odds of a project being extremely delayed. Our findings suggest that cities could expedite transit-accessible housing development by ensuring that general plans and zoning accommodate multifamily development near transit.
The transportation sector is a major source of California’s greenhouse gas emissions, contributing 41% of the state total[1]. California policy is moving rapidly toward Zero Emission battery electric vehicles (BEV) and hydrogen fuel cell vehicles (FCV). Governor Newsom has issued an executive order that all new in-state sales of passenger vehicles should be Zero Emission Vehicles (ZEV) by 2035. Further, the California Air Resources Board has approved rulemaking requiring that more than half of trucks sold in the state must be zero-emissions by 2035, and all of them by 2045 [1a].California has the ambitious goal of achieving a 60% renewable electricity grid by 2030 and 100% carbon free grid by 2045. High penetration of variable renewable energy (VRE) requires seasonal storage to match supply and demand and hydrogen could be a possible candidate for this purpose [1b]. The author has developed the CALZEEV energy-economic model to study possible roles for hydrogen in a VRE intensive future grid with a large Zero Emission Vehicle fleet, comprised of both BEVs and FCVs. In particular, we study whether we can provide sufficient seasonal storage for a 100% zero carbon electricity grid and the potential role of H2 infrastructure in a BEV/FCEV combination for a sustainable path towards a zero-emission energy system. The role of hydrogen infrastructure in seasonal storage for balancing VRE generation while meeting demand for hydrogen vehicles year around has been studied, including economic impacts.
