research report

Pattern Recognition for Curb Usage

Abstract

The increasing use of transportation network companies and delivery services has transformed the utilization of curb space, resulting in a lack of parking and contributing to congestion. No systematic method exists for identifying curb usage patterns, but emerging machine learning technologies and low-tech data sources, such as dashboard cameras mounted on vehicles that routinely travel the area, have the potential to monitor curb usage. To demonstrate how video data can be used to recognize usage patterns, the research team conducted a case study on Bancroft Way in Berkeley, CA. The project collected video footage with GPS data from a dashboard camera installed on a shuttle bus that circles the area. The team trained a machine learning model to recognize different types of delivery vehicles in the data images and then used the model to visualize curbside usage trends. The findings include identifying hot spots, analyzing arrival patterns by delivery vehicle type, detecting bus lane blockage, and assessing the impact of parking on traffic flow. The proof-of-concept study demonstrated that machine learning techniques when coupled with affordable hardware like a dashboard camera, can reveal curb usage patterns. The data can be used to efficiently manage curb space, facilitate goods movement, improve traffic flow, and enhance safety.

conference paper

Heterogeneity in Activity-travel Patterns of Public Transit Users

Abstract

Public transit is considered a sustainable mode of transport that can address automobile dependency and provide environmental, economic, and societal benefits. However, with typical temporal and spatial constraints such as fixed routes and schedules, transfer requirements, waiting times, and access/egress issues, public transit offers lower accessibility and mobility services than private vehicles and thus it is considered a less attractive mode to many prospective users. To improve the performance of transit and in turn to increase its usage, a broader understanding of the daily activity-travel patterns of transit users is fundamental. In this context, this study analyzed transit-based activity-travel patterns by classifying users via Latent Class Analysis (LCA). Using data from the 2017 National Household Travel Survey, the LCA model suggested that transit users could be split into five distinct classes where each class has a representative activity-travel pattern. Class 1 constituted employed white males who made transit-dominant simple work tours. Class 2 was composed of employed white females who made complex work tours. Employed white millennials comprised Class 3 and made multimodal complex tours. Transit Class 4 were non-white younger or older adult groups who made transit-dominant simple non-work tours. Last, Class 5 members made complex non-work tours with recurrent transit use and comprised single older women. This study provided insights regarding the variations of activity-travel patterns and the associated market segments of transit users in the United States. The results can assist transit agencies in identifying transit user groups with particular activity patterns and considering market strategies that can address their travel needs.

research report

Testing Wildfire Evacuation Strategies and coordination plans for Wildland-Urban Interface (WUI) Communities in California

Publication Date

April 1, 2024

Author(s)

Kenichi Soga, Louise Comfort, Bingyu Zhao, Pengshun Li, Paola Lorusso

Abstract

In the event of a wildfire, government agencies need to make quick, well-informed decisions to safely evacuate people. Small communities, such as in Marin County, with a mix of residences and flammable vegetation in Wildland-Urban Interface zones, tend to lack resources to conduct evacuation studies. Consequently, this study uses a framework of wildfire and traffic simulations to test the performance of potential evacuation strategies, including reducing the volume of evacuating vehicles through car-pooling, phasing evacuations by staggering evacuation times by zone, and prohibiting street parking in four representative areas of Marin County. Results show that reducing vehicle numbers lowers the average travel time by 20%-70% and average exposure time to wildfire by 27%-60% from the baseline. Phased evacuations with suitable time intervals lower the average travel time by 13.5%-70% but may expose more vehicles to fire in some situations. Prohibiting street parking yields varying results due to different numbers of exits and evacuees. In some cases, prohibiting street parking reduces the average travel time by over 50%, while in other cases it only reduces the average travel time by 9%, contributing little to evacuation efficiency. Altogether, Marin County may want to consider developing a communication and parking plan to reduce the number of evacuating vehicles in wildfire situations. Phased evacuation is also highly recommended, but the suitable phasing interval depends on the speed of fire spread and number of evacuees. Further, whether to establish street parking prohibition policies for a certain area depends on the number of exits and the number of vehicles on the streets.

policy brief

A Case Study: Testing Wildfire Evacuation Strategies for Communities in Marin County, California

Publication Date

April 1, 2024

Author(s)

Kenichi Soga, Louise Comfort, Bingyu Zhao, Pengshun Li, Paola Lorusso

Abstract

Many small, resource-strapped communities located in areas vulnerable to wildfire don’t have resources to conduct dedicated evacuation studies and many do not consider the impact of background traffic (i.e., normal traffic rather than evacuating traffic) on evacuation. In response, the researchers explored the performance of several generalizable evacuation strategies with background traffic for representative communities in Marin County, including the Ross Valley, Woodacre Bowl, Tamalpais Valley, and an area near Highway 101 and Ignacio Boulevard in Novato (hereafter referred to as ‘Novato Neighborhood’). The strategies explored include vehicle reduction (i.e., evacuees share a vehicle), phased evacuation (i.e., evacuees in different zones have different departure times), and off-street parking (i.e., street parking is prohibited on a high-fire Red Flag Day to increase overall road capacity in the event of an evacuation). The researchers then tested each strategy using a wildfire-traffic simulation framework.

policy brief

Higher Bus Ridership Unlikely to Increase Community COVID-19 Transmission

Abstract

Public transportation has been blamed for facilitating the spread of COVID-19 in dense, urban areas. As a response to the COVID-19 pandemic, transit agencies have reduced service and implemented mask-wearing mandates and social distancing aboard transit. Some prior studies that address public transportation provide some evidence that negative COVID-19 outcomes are linked to high transit use. One early study of COVID-19 transmission on trains in China found that transmission is also affected by the density of passengers, seat spacing, and length of time traveled with other passengers aboard the trains.

policy brief

What Drives Shared Micromobility Ridership?

Abstract

Shared micromobility (e.g., e-scooters, bikes, e-bikes) offers moderate-speed, space-efficient, and “carbon-light” mobility, promoting environmental sustainability and healthy travel. While the popularity and use of shared micromobility has grown significantly over the past decade, it represents a small share of total trips in urban areas. To better understand shared micromobility ridership, researchers from across the U.S. and the world have analyzed statistical associations between shared micromobility usage and various explanatory factors, including socio-demographic and -economic attributes, land use and built environment characteristics, surrounding transportation options (e.g., public transit stations), geography (e.g., elevation), and micromobility system characteristics (e.g., station capacity). To understand what these studies collectively mean in terms of expanding shared micromobility usage, we conducted a meta-analysis of 30 empirical studies and then developed robust estimates of factors that encourage ridership across different markets.

published journal article

Important Considerations in Machine Learning-based Landslide Susceptibility Assessment Under Future Climate Conditions

Abstract

Rainfall-induced landslides have caused a large amount of economic losses and casualties over the years. Machine learning techniques have been widely applied in recent years to assess landslide susceptibility over regions of interest. However, a number of challenges limit the reliability and performance of machine learning-based landslide models. In particular, class imbalance in the dataset, selection of landslide conditioning factors, and potential extrapolation problems for landslide prediction under future conditions need to be carefully addressed. This work introduces methodologies to address these challenges using XGBoost to train the landslide prediction model. Data resampling techniques were adopted to improve the model performance with the imbalanced dataset. Various models were trained and their performances evaluated using a combination of different metrics. The results show that synthetic minority oversampling technique combined with the proposed gridded hyperspace sampling technique performs better than the other imbalance learning techniques with XGBoost. Subsequently, the extrapolation performance of the XGBoost model was evaluated, showing that the predictions remain valid for the projected climate conditions. As a case study, landslide susceptibility maps in California were generated using the developed model and compared with the historical California landslide catalog. These results suggest that the developed model can be of great significance in global landslide susceptibility mapping under climate change scenarios.

published journal article

How do they get by without cars? An analysis of travel characteristics of carless households in California

Abstract

In spite of their substantial number in the U.S., our understanding of the travel behavior of households who do not own motor vehicles (labeled “carless” herein) is sketchy. The goal of this paper is to start filling this gap for California. We perform parametric and non-parametric tests to analyze trip data from the 2012 California Household Travel Survey (CHTS) after classifying carless households as voluntarily carless, involuntarily carless, or unclassifiable based on a CHTS question that inquires why a carless household does not own any motor vehicle. We find substantial differences between our different categories of carless households. Compared to their voluntarily carless peers, involuntarily carless households travel less frequently, their trips are longer and they take more time, partly because their environment is not as well adapted to their needs. They also walk/bike less, depend more on transit, and when they travel by motor vehicle, occupancy is typically higher. Their median travel time is longer, but remarkably, it is similar for voluntarily carless and motorized households. Overall, involuntarily carless households are less mobile, which may contribute to a more isolated lifestyle with a lower degree of well-being. Compared to motorized households, carless households rely a lot less on motor vehicles and much more on transit, walking, and biking. They also take less than half as many trips and their median trip distance is less than half as short. This study is a first step toward better understanding the transportation patterns of carless households.

preprint journal article

Predicting Vehicular Emissions by Converging Direct Measurements and Mobility Science

Abstract

Vehicle emissions pose a significant challenge for cities worldwide, yet a comprehensive analysis of the relationship between mobility metrics and total vehicle emissions at a high resolution remains elusive. In this work, we introduce the Mobile Data Emission System (MODES), a pioneering framework that integrates various sources of individual mobility data on an unprecedented scale. Our model is validated with direct measurements from a network of high-density sensors analyzed before and during the COVID-19 pandemic shelter-in-place orders. MODES is used as a laboratory for scaling analysis. Informed by individual trips, we estimate the traffic CO2 emissions at a metropolitan scale with a combination of 3 accessible metrics: vehicle kilometers traveled (VKT), congestion levels, and vehicle efficiency. Given their ranges of variation, VKT has the greatest role in amplifying vehicular emissions up to 500%, followed by vehicle efficiency that would range from 20% to 300% of the average passenger combustion vehicles. In comparison, congestion amplifies vehicle emissions of individual travels by up to 50%. We confirm that cities in the Bay Area with high population density are consistently characterized by low per-person VKT. Nevertheless, high population density comes at the expense of increased congestion. Since VKT is the governing factor, overall densifying of the urban landscape reduces transportation emissions despite its impacts on congestion.

published journal article

Mobile Phone Location Data for Disasters: A Review from Natural Hazards and Epidemics

Abstract

Rapid urbanization and climate change trends, intertwined with complex interactions of various social, economic, and political factors, have resulted in an increase in the frequency and intensity of disaster events. While regions around the world face urgent demands to prepare for, respond to, and recover from such disasters, large-scale location data collected from mobile phone devices have opened up novel approaches to tackle these challenges. Mobile phone location data have enabled us to observe, estimate, and model human mobility dynamics at an unprecedented spatio-temporal granularity and scale. The COVID-19 pandemic, in particular, has spurred the use of mobile phone location data for pandemic and disaster management. However, there is a lack of a comprehensive review that synthesizes the last decade of work and case studies leveraging mobile phone location data for response to and recovery from natural hazards and epidemics. We address this gap by summarizing the existing work and point to promising areas and future challenges for using mobile phone location data to support disaster response and recovery.