Ashour, L., & Shen, Q. (2025). Unveiling post-pandemic commute choices amidst the rise of telework. Transportation Research. Part D, Transport and Environment, 149, Article 105068. https://doi.org/10.1016/j.trd.2025.105068
View Publication
Abstract
The post-pandemic “new normal” is characterized by increased adoption of telework and a decline in transit use, with a growing uncertainty about the long-term effects of these trends. This study investigates the impact of telework adoption on commute mode choice and the factors influencing individual preferences in the new normal, using data from the 2022 Seattle Commute Survey, capturing mode choice on each day in a typical workweek, providing a mixture of modes and the frequency of use for each respondent. Using the Multiple Discrete-Continuous Extreme Value (MDCEV), the study provides timely insights into the factors influencing commute mode choice, highlighting the significant influence of sociodemographic, job-related, and built environment characteristics on mode choice and the nuanced difference between modes, raising concerns about potential inequities in the new normal. The findings highlight the need to prioritize access to public transit, adapt transportation infrastructure to telework, and address rising inequities.
Sa, H., & Shen, Q. (2026). Understanding employees’ residential choices in connection to remote work during the COVID-19 pandemic. Cities, 173, Article 107008. https://doi.org/10.1016/j.cities.2026.107008.
View Publication
Abstract
The outbreak of COVID-19 resulted in the proliferation of work-from-home (WFH) and flexible work arrangements. Existing studies have not systematically investigated the resulting patterns of employees' residential choices within the metropolitan area. This study aims to deepen the understanding of employees' residential location choices during the COVID-19 pandemic by examining the impacts of their remote work status. The study is based on the 2022 Seattle Commute Survey, which collected information on work arrangements, individuals' sociodemographic characteristics, work and home locations, and commute and non-commute travel. The primary results of our analysis indicated that: (1) relocated employees who worked remotely (full WFH and hybrid) were more inclined to move outward from their work locations, compared to those engaged in in-person work, (2) employees relocating closer had the shortest commute distances across all work arrangements, while those relocating farther experienced the longest commutes, especially among full WFH employees, (3) remote work status (full WFH and hybrid) had a strong positive association with the choices of staying in original homes and relocating farther from worksites compared to the reference choice of relocating closer to worksites, (4) the influences of full WFH and hybrid work arrangements on the choice of staying in the original housing and relocating farther did not generally differ across age or household income groups, and (5) residential decisions were jointly shaped by a combination of work policy, sociodemographic elements, household lifecycle stage, commuting characteristics, and neighborhood attributes. Our findings contribute to the literature on employees' intra-metropolitan residential choices in connection with teleworking.
Keywords
Residential choices; Remote work; COVID-19 pandemic; Bayesian multilevel multinomial logistic model; Urban transportation planning
Ling, C., Chen, P., Chu, Y.-C., & Shen, Q. (2026). Not all teleworkers reduce travel: An intensity-dependent behavioral framework with explainable machine learning. Transport Policy, 187, Article 104267. https://doi.org/10.1016/j.tranpol.2026.104267
View Publication
Abstract
Previous studies have yielded mixed findings on the relationship between teleworking and travel, possibly due to overlooking teleworking intensity and its evolving impacts. This study examines how teleworking is associated with overall vehicle miles traveled (VMT) and vehicle travel time (VTT) across a spectrum of intensities, measured by the number of remote-work days per week. Using travel survey data in the Puget Sound Region, Washington, we estimate four explainable machine learning models (XGBoost with SHAP) that incorporate behavioral, socio-demographic, and built-environmental characteristics. Results find an intensity-dependent relationship among typical workers. Low-intensity teleworking (1 day/week) is associated with increased travel both pre- and post-pandemic, while moderate teleworking (2 days/week) shows near-neutral or marginally positive associations. In contrast, regular teleworking (3–5 days/week) is correlated with reduced travel. In 2023, teleworking intensity ranks among the top six contributors, accounting for 6.84% and 4.08% of model explainability for VMT and VTT, respectively, higher than in 2017. The reductions scale with intensity. Holding other factors constant, high-frequency teleworkers (3 days/week) in 2023 accrued approximately 10 fewer miles and 3 fewer minutes per five-day workweek than fully on-site workers, whereas full teleworkers (5 days/week) saved 19 fewer miles and 23 fewer minutes. These impacts are concentrated among high-frequency teleworkers with longer commute distances, while the effects are mixed among shorter-distance commuters. Teleworking can moderate impacts of other factors, such as commute distance, income, and population density. These findings help reconcile prior inconsistencies, showing that teleworking's role depends on intensity rather than simple adoption.
Research Interests: Machine learning, work from home, urban economics, transportation policy, urban data science, economic inequality
Research Interests: Last-mile delivery and freight sustainability
Ashour, L., & Shen, Q. (2025). An outlook on ride-sourcing price changes: Implications for future transit agency-TNC partnerships. Transport Policy, 173, Article 103790. https://doi.org/10.1016/j.tranpol.2025.103790
View Publication
Abstract
Ride-sourcing trip prices charged by transportation network companies (TNCs) have increased significantly compared to before the pandemic, causing concerns about the effectiveness of existing and planned transit agency-TNC partnerships. This paper explores three scenarios of future TNC price changes: (1) price trend extension using forecasting models, (2) price increase in response to local policy changes, and (3) TNC/taxi price convergence due to increased competition. We then investigate the impact of TNC price change on the prospect of transit agency-TNC partnerships, using a case study in the Seattle region. For the first scenario, we employ two time-series models, namely ARIMA and PROPHET, to forecast price changes within the next three years (Oct 2022–Oct 2025) using publicly available Chicago TNC trip data. The results show that TNC's daily average price would reach $3.23 per mile, increasing by 40 % from 2019 average rates. For the second scenario, we track significant policies that directly impacted TNC prices in Seattle and incorporate reported price increases. The resulting estimations indicate that TNC prices would increase by an extra 25 % in response to changes in the minimum wage law. For the third scenario, we use publicly available taxi trip data of the city of Chicago and forecast future taxi prices by estimating time-series models comparable to those for TNC prices. The analysis suggests that due to increased competition, TNC and taxi prices are converging and that the average TNC fare per mile could add another 50 % to the forecasted price if TNC and taxi prices become similar in the upcoming three years. These price changes are shown to have a considerable negative impact on the expected cost-effectiveness of transit agency-TNC partnerships. Although such partnerships could still provide many benefits, transportation planners and policymakers should carefully examine the implications of TNC price increases resulting from changing market and policy environments.
Ashour, L., & Shen, Q. (2025). Incorporating equity into the cost-effectiveness evaluation of new mobility: A comparative analysis. Transportation Research. Part D, Transport and Environment, 147, Article 104959. https://doi.org/10.1016/j.trd.2025.104959
View Publication
Abstract
Public transportation in suburban areas faces challenges in providing efficient mobility. Transit Incorporating Mobility on Demand (TIMOD) services have emerged as a potential solution, yet equity considerations remain underexplored. This study incorporates equity into the cost-effectiveness evaluation of TIMOD services, analyzing two suburban areas in the Seattle metropolitan region where a TIMOD service is implemented. Using distributional cost-effectiveness analysis (DCEA), we assess the comparative costs of TIMOD, fixed-route transit, and drive-alone across different income groups and built environments. The study shows that although TIMOD services offer equity benefits for lower-income travelers, they are more equitable in high-density, low-income suburbs. In contrast, their cost-effectiveness is more limited in affluent, low-density areas. These insights highlight the importance of context-specific planning for TIMOD interventions and employ tools such as DCEA for transit agencies to prioritize the deployment of such services in areas where they can maximize social welfare and reduce transportation inequities.
Cai, M., Ashour, L. A., Shen, Q., & Chen, C. (2025). Incorporating mobility-on-demand into public transit in suburban areas: A comparative cost-effectiveness evaluation. Transportation Research. Part D, Transport and Environment, 144, Article 104775. https://doi.org/10.1016/j.trd.2025.104775
View Publication
Abstract
Transit Incorporating Mobility-on-Demand (TIMOD) represents the public–private partnerships in which transit agencies incorporate MOD services to supplement fixed-route transit. This study evaluates the cost-effectiveness of TIMOD compared to buses, driving, and ride-hailing in suburban settings. For each alternative, it estimates the marginal costs for travelers, service providers, and transportation externalities, which constitute the marginal social cost. In the study cases, TIMOD is the least cost-effective option, with marginal social cost approximately 20% higher than TNCs and over three times higher than driving. For travelers, TIMOD costs more than driving but less than buses and ride-hailing when considering time value and fare. The cost of TIMOD declines as population density increases. Suburbs with less bus services and higher income residents benefit more from TIMOD, realizing greater reductions in time costs compared to buses. Transit agencies should explore alternative ways to improve mobility for disadvantaged suburban residents by offsetting driving costs and subsidizing TNCs fares.
Keywords
Transit Incorporating Mobility-On-Demand (TIMOD); Public transit; Transportation simulation; Suburban areas; Marginal social cost of travel
Cai, M., Shen, Q., Wang, Y., Brown, M., Ban, X., & Ashour, L. A. (2024). Examining commute mode choice of essential workers before and during the COVID-19 pandemic – A case study of the University of Washington. Case Studies on Transport Policy, 15, 101129-. https://doi.org/10.1016/j.cstp.2023.101129
View Publication
Abstract
During disruptive events and major crises such as the COVID-19 pandemic, essential workers, defined in this study as employees who must continue to make frequent commute trips, play a critical role in maintaining the basic function of society. It is therefore vital for transportation planners and policymakers to ensure that adequate mobility services are provided to meet the travel needs of this group of workers. To better understand essential workers' mobility barriers and inform future transportation policy, this research examines their pre-and during-pandemic commutes, as well as prospects on future commuting, using data from the Transportation Needs Assessment Survey conducted by the University of Washington in Seattle. The results show that, first, over 60% of pre-pandemic public transit riders switched to other modes, especially driving alone. In contrast, almost all the essential workers who drove alone, biked, or walked before the pandemic continued to do so during the pandemic. Second, the shift to driving alone was most pronounced among essential workers with high incomes, whereas public transit remained as a primary mode choice of lower-income groups. As travel distance increased, the probability of driving alone over public transit also went up, although the relationship was not linear. We did not find consistently significant associations between mode choice and other sociodemographic variables. Finally, our analysis indicated that most essential workers anticipated to return to their pre-pandemic commute frequency and mode choice when the pandemic is under control. Transportation planners and policymakers can facilitate mobility for essential workers through planning and policy innovations, especially during disruptive events and major crises.
Keywords
Essential workers; Transportation equity; Mode choice; Commuting; COVID-19
Cai, M., Acolin, A., Moudon, A. V., & Shen, Q. (2023). Developing a multi-criteria prioritization tool to catalyze TOD on publicly owned land areas. Cities, 143, 104606-. https://doi.org/10.1016/j.cities.2023.104606
View Publication
Abstract
Public agencies can take a leading role in catalyzing TOD by making land available to developers (selling or leasing land, potentially below market prices). In particular, park-and-ride areas that are publicly owned can be leveraged to support TOD uses, such as affordable housing, office space, small businesses, and mixed-use buildings given their convenient access to transit systems and often large land areas. However, few previous studies have discussed the use of publicly owned park-and-rides, which are an important component of publicly owned land, as a catalyst for TOD. To fill the gap in the literature and effectively support TOD planning, this research developed a multi-criteria prioritization tool to identify the most promising locations for TOD and tested it at three park-and-ride sites owned by the Washington State Department of Transportation. The tool was developed through the Delphi process, which is an effective and inexpensive approach to evaluate relevant indicators by synthesizing the opinions of experts from various backgrounds. Five categories with a total of 14 TOD indicators, including transit supportive land-use zoning, job accessibility, land price, land-use mix, and household income, were selected as measures of TOD suitability. The importance of these indicators varied with three different TOD scenarios: (1) emphasis on affordable housing, (2) emphasis on market-rate housing, and (3) emphasis on mixed-use development. Using the calculated suitability scores, this tool can prioritize potential TOD sites for further review.
Keywords
TOD; Delphi method; Multi-criteria planning tool; Multi-sources geospatial data; Publicly owned land