Talks and presentations

Do racial and ethnic disparities in following stay-at-home orders influence COVID-19 health outcomes? A mediation analysis approach

February 01, 2022

Talk, 2021 Transportation Research Board's 101th Annual Meeting (TRB), Washington D.C., U.S.

Abstract: Racial/ethnic disparities are among the top-selective underlying determinants associated with the disproportional impact of the COVID-19 pandemic on human mobility and health outcomes. This study jointly examined county-level racial/ethnic differences in compliance with stay-at-home orders and COVID-19 health outcomes during 2020, leveraging two-year geo-tracking data of mobile devices across ~4.4 million point-of-interests (POIs) in the contiguous United States. Through a set of structural equation modeling, this study quantified how racial/ethnic differences in following stay-at-home orders could mediate COVID-19 health outcomes, controlling for state effects, socioeconomics, demographics, occupation, and partisanship. Results showed that counties with higher Asian populations decreased most in their travel, both in terms of reducing their overall POIs’ visiting and increasing their staying home percentage. Moreover, counties with higher White populations experienced the lowest infection rate, while counties with higher African American populations presented the highest case-fatality ratio. Additionally, control variables, particularly partisanship, median household income, percentage of elders, and urbanization, significantly accounted for the county differences in human mobility and COVID-19 health outcomes. Mediation analyses further revealed that human mobility only statistically influenced infection rate but not case-fatality ratio, and such mediation effects varied substantially among racial/ethnic compositions. Last, robustness check of racial gradient at census block group level documented consistent associations but greater magnitude. Taken together, these findings suggest that US residents’ responses to COVID-19 are subject to an entrenched and consequential racial/ethnic divide.

Examining spatiotemporal evolution of racial/ethnic disparities in human mobility and COVID-19 health outcomes: Evidence from the contiguous United States

February 01, 2022

Talk, 2021 Transportation Research Board's 101th Annual Meeting (TRB), Washington D.C., U.S.

Abstract: Social distancing has become a key countermeasure to contain the dissemination of COVID-19. This study examined county-level racial/ethnic disparities in human mobility and COVID-19 health outcomes during the year 2020 by leveraging geo-tracking data across the contiguous US. Sets of generalized additive models were fitted under cross-sectional and time-varying settings, with percentage of mobility change, percentage of staying home, COVID-19 infection rate, and case-fatality ratio as dependent variables, respectively. After adjusting for spatial effects, built environment, socioeconomics, demographics, and partisanship, we found counties with higher Asian populations decreased most in travel, counties with higher White and Asian populations experienced the least infection rate, and counties with higher African American populations presented the highest case-fatality ratio. Control variables, particularly partisanship and education attainment, significantly influenced modeling results. Time-varying analyses further suggested racial differences in human mobility varied dramatically at the beginning but remained stable during the pandemic, while racial differences in COVID-19 outcomes broadly decreased over time. All conclusions hold robust with different aggregation units or model specifications. Altogether, our analyses shine a spotlight on the entrenched racial segregation in the US as well as how it may influence the mobility patterns, urban forms, and health disparities during the COVID-19.

An Interactive Covid-19 Mobility Impact and Social Distancing Analysis Platform

February 01, 2021

Talk, 2020 Transportation Research Board's 100th Annual Meeting (TRB), Washington D.C., U.S.

Abstract: The research team has utilized privacy-protected mobile device location data, integrated with COVID-19 case data and census population data, to produce a COVID-19 impact analysis platform that can inform users about the effects of COVID-19 spread and government orders on mobility and social distancing. The platform is being updated daily, to continuously inform decision-makers about the impacts of COVID-19 on their communities using an interactive analytical tool. The research team has processed anonymized mobile device location data to identify trips and produced a set of variables including social distancing index, percentage of people staying at home, visits to work and non-work locations, out-of-town trips, and trip distance. The results are aggregated to county and state levels to protect privacy and scaled to the entire population of each county and state. The research team are making their data and findings, which are updated daily and go back to January 1, 2020, for benchmarking, available to the public in order to help public officials make informed decisions. This paper presents a summary of the platform and describes the methodology used to process data and produce the platform metrics.

Mobile Device Location Data Reveals Human Mobility Response to Stay-at-Home Orders during the COVID-19 Pandemic in the U.S.

February 01, 2021

Talk, 2020 Transportation Research Board's 100th Annual Meeting (TRB), Washington D.C., U.S.

Abstract: One approach to delay the spread of the novel coronavirus (COVID-19) is to reduce human travel by imposing travel restriction policies. Understanding the actual human mobility response to such policies remains a challenge due to the lack of ground truth and large-scale dataset describing human mobility during the pandemic. This study utilizes an integrated dataset, consisting of anonymized and privacy-protected location data that covers over 150 million monthly active samples in the U.S., COVID-19 case data, and census population information, to uncover mobility changes during COVID-19 and under the “Stay-at-home” state orders in the U.S. The study successfully quantifies human mobility responses with three important metrics: daily average number of trips per person; daily average person-miles traveled; and daily percentage of residents staying home. The data analytics reveal a voluntary mobility reduction that occurred regardless of government actions, and a “floor” phenomenon that human mobility reached a lower bound and stopped decreasing soon after each state announced the “Stay-at-home” order. A set of longitudinal models is then developed and confirms empirically that about 5% of the reduction in human mobility is due to the effect of states’ “Stay-at-home” policy. Lessons learned from the data analytics and longitudinal models offer valuable insights for government actions in preparation for another COVID-19 or other virus outbreak in the future.

A Big-Data Driven Approach to Analyzing and Modeling Human Mobility Trend under Non-Pharmaceutical Interventions during COVID-19 Pandemic

February 01, 2021

Talk, 2020 Transportation Research Board's 100th Annual Meeting (TRB), Washington D.C., U.S.

Abstract: During the unprecedented coronavirus disease 2019 (COVID-19) challenge, non-pharmaceutical interventions became a widely adopted strategy to limit physical movements and interactions to mitigate virus transmissions. For situational awareness and decision-support, quickly available yet accurate big-data analytics about human mobility and social distancing is invaluable to agencies and decision-makers. This paper presents a big-data-driven analytical framework that ingests terabytes of data daily and quantitatively assesses the human mobility trend during COVID-19. Using mobile device location data of over 100 million monthly active samples in the United States, the study successfully measures human mobility with three main metrics at the county level: daily average number of trips per person; daily average person-miles traveled; and daily percentage of residents staying home. A set of generalized additive mixed models is employed to disentangle the policy effect on human mobility from other confounding effects including virus effect, socio-demographic effect, weather effect, and spatiotemporal autocorrelation. Results reveal the policy plays a limited, time-decreasing, and region-specific effect on human movement. The stay-at-home orders only contribute to a 3%-7.3% decrease in human mobility, while the reopening guidelines lead to a 1%-4.7% mobility increase. Results also indicate a reasonable spatial heterogeneity among the U.S. counties, wherein the number of confirmed COVID-19 cases, income levels, age and racial distribution play important roles. The data informatics generated by the framework are made available to the public for a timely understanding of mobility trends and policy effects, as well as for time-sensitive decision support to further contain the spread of the virus.

Exploring the effect of battery capacity on electric vehicle sharing programs using a simulation approach

February 01, 2020

Talk, 2019 Transportation Research Board's 99th Annual Meeting (TRB), Washington D.C., U.S.

Abstract: Aided by mobile computing technology, shared electric vehicles (SEVs) have become an accessible and affordable mobility option. However, limited battery capacity remains a major obstacle for large-scale adoption of SEVs, and greatly undermines their popularity. In this study, a discrete-event simulation approach was employed to estimate how battery capacity affects the performance of a carsharing program. Results show that limited battery capacity lowered user satisfaction and vehicle utilization in the program. Increased charging speed, maximum range, and vehicle-to-trip ratio help mitigate these negative effects. Specifically, increasing the maximum range or charging speed contributes to the increment of the average SEV usage time and the percentage of satisfied rental requests. A higher vehicle-to-trip ratio contributes to a greater level of user satisfaction but a lower level of vehicle utilization. Additionally, the negative effects of battery capacity are greatly diminished after charging speed is increased to a certain threshold. These findings help capture the trade-off between charging facility investment, vehicle utilization, and user satisfaction. Increasing charging speed and maximum range are necessary if operators want to maximize vehicle utilization and promote user satisfaction. However, this investment must also account for cost-effectiveness.

Do Personalized Economic Incentives Work in Promoting Shared Mobility? Examining Customer Churn Using a Time-Varying Cox Model

February 01, 2020

Talk, 2019 Transportation Research Board's 99th Annual Meeting (TRB), Washington D.C., U.S.

Abstract: Shared mobility options have grown significantly in recent years. However, a major threat to the growing usage of shared-cars is customer churn. Despite companies continually providing rewards to attract customers, factors that contribute to customer churn are not well-understood in the existing literature. This study analyzed one-year transaction data of a one-way carsharing program and applied an extended Cox proportional hazards model to examine these effects. Our results show that: (1) A customer’s survival probability decreased over time with a decelerating rate. (2) Customers were less likely to churn when their frequently-visited shared-car stations were located in neighborhoods with low access to transit or in areas with colleges, railway hubs or airports. (3) Older and male customers had a lower likelihood of churning. (4) The number of coupons was negatively associated with the likelihood of customer churn. (5) The effect of coupons generally increased at first and then decreased after the 130th day. These findings contribute to coupon design and the location choice of shared-car stations. Specifically, we suggest that large-denomination coupons should be split into multiple small-denomination coupons. We encourage companies to locate stations in areas with poor transit access. Meanwhile, coupon issuing strategies can be redeveloped. Offering more coupons to new customers can help maximize the profits of service providers. However, such a strategy is debatable because it invades personal privacy and is unfair for older, loyal customers.

Estimating traffic volume for local streets with imbalanced data

February 01, 2019

Talk, 2018 Transportation Research Board's 98th Annual Meeting (TRB), Washington D.C., U.S.

Abstract: Annual average daily traffic (AADT) is an important measurement used in traffic engineering. Local streets are major components of a road network. However, automatic traffic recorders (ATRs) used to collect AADT are often limited to arterial roads, and such information is, therefore, often unavailable for local streets. Estimating AADT on local streets becomes a necessity as local street traffic continues to grow and the capacity of arterial roads becomes insufficient. A challenge is that an under-represented sample of local street AADT may result in biased estimation. A synthetic minority oversampling technique (SMOTE) is applied to oversample local streets to correct the imbalanced sampling among different road types. A generalized linear mixed model (GLMM) is employed to estimate AADT incorporating various independent variables, including factors of roadway design, socio-demographics, and land use. The model is examined with an AADT dataset from Seattle, WA. Results show that: (1) SMOTE helps to correct imbalanced sampling proportions and improve model performance significantly; (2) the number of lanes and the number of crosswalks are both positively associated with AADT; (3) road segments located in areas with a higher population density or more mixed land use have a higher AADT; (4) distance to the nearest arterial road is negatively correlated with AADT; and (5) AADT creates spatial spillover effects on neighboring road segments. The combination of SMOTE and GLMM improves the estimation accuracy on AADT, which contributes to better data for transportation planning and traffic monitoring, and to cost saving on data collection.

Impacts of Rain and Waterlogging on Traffic Speed and Volume on Urban Roads

October 01, 2018

Talk, 2018 21st International Conference on Intelligent Transportation Systems (ITSC), Hawaii, U.S.

Abstract: Rain causes significant drops in traffic speed and volume. Heavy rain can induce road waterlogging especially in the developing countries due to their poor drainage systems, which will exacerbate its effect on transportation systems. 45 urban waterlogging segments in Shenzhen, China are examined in this paper to quantify the effects of rainfall and rain-induced waterlogging on the traffic speed and volume using both descriptive and statistical methods. Results indicate that drops in travel speed and volume caused by waterlogging (-21.5 % to -24.3%, -25.8% to -31.2% respectively) is generally greater than rain (-1.2% to -18.4%, -1.1% to -16.5% respectively). It is also found that reductions in volume during the off-peak period are generally greater than those during peak hours however an opposite pattern is observed for speed. This research provides the intelligent transportation system with detailed information which is helpful for early warning, risk management and weather-responsive transportation management strategies devisal under rainy and waterlogging conditions. Further research might focus on the connections between research findings and their applications in the ITS context.

Modeling users’ vehicles selection behavior in the urban carsharing program

October 01, 2018

Talk, 2018 21st International Conference on Intelligent Transportation Systems (ITSC), Hawaii, U.S.

Abstract: Electric carsharing network has been expanding at a very fast rate in the last few years, accompanied by more complex challenges to carsharing operators. Increasing vehicles usage allows for benefits maximization. However, vehicles usage varies significantly in a fleet because of users’ preference to vehicles with different features. This study investigated contributing factors to users’ vehicles selection behavior through random forests and binary logistic regression using the administrative datasets collected from EVCARD carsharing program. Results showed state of charge (SOC) of electric vehicles and the number of available vehicles parked at a station had the greatest effect on user’s vehicles selection behavior. Users tend to be greedy rather than rational when making decisions as they always choose the vehicles with the maximum SOC even their real trips are short. The attributes of trips and users, like real trip distances and users’ familiarity with carsharing program, also play an important role in the selection process. Findings from this research can be beneficial to carsharing operators to prioritize investments when purchasing new vehicles and develop optimal management strategies to enhance existing vehicles attraction for users.

Songhua Hu

Talks and presentations