Activity-Based Model Application for Business Reopening Scenarios after COVID-19
Vladimir Livshits, Maricopa Association of GovernmentsShow Abstract
Arup Dutta, Maricopa Association of Governments
Petya Maneva, Maricopa Association of Governments
Kyunghwi Jeon, Maricopa Association of Governments
Daehyun You, Maricopa Association of Governments
Haidong Zhu, Maricopa Association of Governments
Peter Vovsha, INRO
Gaurav Vyas, INRO
James Hicks, WSP
David Ory, WSP
COVID-19 made a profound impact on all aspects of life and associated activity-travel behavior. The primary purpose of the maximum business closure, except for the essential businesses, in the period of March-April 2020 was saving human lives and stopping the virus from spreading, especially in urbanized areas. At the current stage starting from May 2020, a gradual reopening of businesses is taking place. Business reopening scenarios should be planned and staged to properly balance the socio-economic benefits with the still remaining risks of a further spread of the virus and specifically, affecting the most vulnerable populations such as elderly people. This paper reports an approach to utilize the Activity-Based Model (ABM) recently developed by the Maricopa Association of Governments (MAG) for the purpose of planning and evaluation of busines reopening scenarios in the Phoenix, AZ metropolitan region. The model output can be considered as a “digital twin” of the modelled region where a complete record is available for where people are in the region at any point of time and for what activity. This feature allows for modeling different business reopening scenarios and evaluation of their consequences with regard to a potential further spread of the virus (on one hand) and socio-economic cost/benefit for the regional population and economy (on the other hand).
Capturing the Effects of Social Networks and Mobility on Viral Spread: Agent-based Model and Scenario Application
Hoseb Abkarian, Northwestern UniversityShow Abstract
Ying Chen, Northwestern University
Hani Mahmassani, Northwestern University
The spread of an infectious disease is a highly complex process in modern societies. To capture interactions at a microscopic level, this paper presents DYNASPREAD, a DYNAmic agent-based simulation methodology that incorporates a novel mathematical formulation for a three-phase diffusion process for the SPREAD of a virus. Interactions of individuals are modelled both at a social network level as well as due to random encounters associated with travel activity behavior. The model is calibrated to a small study area in Chicago on COVID-19 data, after which intervention scenarios are tested. Initial results illustrate that enforcement of staying at home is necessary during pandemics: enforcing 60% of people to stay at home (randomly every day) can quickly reduce the number of new infections with 40% reduction in total cases. Benefits of quarantine, on the other hand, substantially relies upon how long people stay sick, asymptomatically. Specifically, if the quarantine duration is short, cases can still dramatically increase after reopening. Finally, teleworking with only short walks allowed shows promise in mitigating the spread of the virus, especially early in the life of the pandemic with near 71% reduction in total cases.
A Segmentation Analysis of the Impacts of COVID-19 on Weekly Travel in the United States: Results from a Longitudinal Survey
Mark Bradley, RSG IncShow Abstract
Abigail Rosenson, RSG
Taylor Daly, RSG Inc
Elizabeth Greene, RSG Inc
To add to the growing body of evidence on COVID-related changes in travel behavior, the study team initiated a longitudinal survey of results across the United States. Wave 1 of the survey was carried out in mid-May 2020, with a final sample size of 3,115 adults, and Wave 2 was carried out in the last week of June and early July 2020, with a further sample of 3,028, adults. The data was collected via a national survey panel, and the data was reweighted to match 2018 Census data. The results indicate that during the first two months of the COVID-related social distancing (March-May 2020) use of motorized modes (auto, transit and paid rideshare) decreased by over 40%, with the largest decreases in the Pacific and Northeast regions where the number of cases was initially highest, and lowest in the Southeast, where the number of COVID cases was initially relative low. Between mid-May and early July, frequency of car use increased enough so that the overall reduction compared to pre-COVID car use was less than 20% in Wave 2, compared to almost 40% in Wave 1. Segmentation analysis indicates that the two factors that have the largest influence on trends in overall car use and mode use is the type of industry that the person works in (with lowest increases in health care, retail manufacturing and construction), and the ability for the person to telecommute from home. These differences further translate into less pronounced differences by race/ethnicity, income, and age group.
Estimating Economic Impacts of COVID-19 Pandemic at the Municipal Level: A Latent Class Regression Modeling Approach
Fariba Hossain, Dalhousie UniversityShow Abstract
Muhammad Habib, Dalhousie University
This study develops an economic model for estimating the loss of sales of business establishments during COVID-19 pandemic in Halifax, Canada. This study considers four different scenarios: business-as-usual, lockdown, conservative and aggressive reopening scenarios. To build the scenarios, Google COVID-19 Mobility Report and Apple Mobility Trend Reports are used. This study utilizes data from an activity-based travel demand model as a source of activity participation by category. A Latent Class Model (LCM) is used to examine relationships between yearly sales and independent variables of interest including business attributes, mobility attributes and built-environment characteristics in the Halifax Regional Municipality (HRM). This model can predict a probable loss of business establishments at the zonal level. The result shows significant effect on the economy that the economy by the pandemic situations. During lockdown, conservative and aggressive scenarios, on average economic loss of 87%, 71% and 60% from business-as-usual scenario is predicted by LCM for the municipality. Spatial distribution of sales within the traffic analysis zones (TAZs) reveals variation of economic impact through TAZs during all pandemic scenarios. Through the phased reopening stages economy started to recover as almost 50% of total TAZs face less than 20% reduction in sales in aggressive scenario. The developed model will be beneficial for the policymakers in determining reopening strategies for future pandemics.
Assessing the Impacts of COVID-19 on Urban Passenger Travel Demand: Description of A Multi-Pronged and Multi-Staged Study with Initial Results
Khandker Nurul Habib (email@example.com), University of TorontoShow Abstract
Jason Hawkins, University of Toronto
Saeed Shakib, University of Toronto
Patrick Loa, University of Toronto
Sk Md Mashrur, University of Toronto
Alireza Dianat, University of Toronto
Kaili Wang, University of Toronto
Sanjana Hossain, University of Toronto
Yicong Liu, University of Toronto
COVID-19 outbreak created a context that was never thought about by any urban transportation planner and modellers. It created a complete halt of urban lifestyle and resulting in a total disruption of travel behaviour. While we are still not out of the cusp of it, it is of great importance that we take a detailed approach to capture the impacts of such disruption with great care. The paper presents a study design to comprehensively measure the effects of COVID-19 induced Lockdown on changes in travel behaviour/demands and its dynamics. The study is designed for the Greater Toronto Area, and the paper presents the results of the common portions of four specialized travel surveys composing a sample of around 4000 survey respondents in the study area. The empirical investigation presented in the paper is on the general pandemic response, in terms of daily activity-travel adaptation behaviour), in the context of around 4 months of complete Lockdown to stop spreading of COVID-19. Empirical investigation reveals that the Lockdown did not cause any substantial increase in unemployment, but it did force the majority of the residents to telecommute and practice flexible office hours. Results of the survey presented in the paper, thus, can be generalized to very large scale, if not 100 percent (as many workers did commute as they needed to be at the workplace) implementation of telecommuting with additional restrictions of out-of-home movements (imposed by the closure of businesses and social distancing). Overall, results show that Lockdown influences people to engage in more family-oriented activities. However, such choices are largely influenced by age, gender, household car ownership, and even the urban form of the cities.
The Effect of Productivity of Telecommuting During the COVID-19 Pandemic on Workers’ Post-pandemic Intentions to Telecommute
Ali Shamshiripour (firstname.lastname@example.org), University of Illinois, ChicagoShow Abstract
Ehsan Rahimi, University of Illinois, Chicago
Ramin Shabanpour, University of North Florida
Nima Golshani, Georgia Institute of Technology (Georgia Tech)
Abolfazl Mohammadian, University of Illinois, Chicago
The COVID-19 pandemic has triggered destructive impacts on various aspects of communities including public health and economy whose consequences are expected to endure for the future. Changing people’s travel behavior, the pandemic has forced people to stay at home and telecommute rather than traveling to their workplaces. Such a rapid shift towards telecommuting among a considerable portion of the employment population gives us a unique opportunity to investigate the future of telecommuting with a focus on the workers productivity while working at home. In particular, we aim to understand what underlying factors motivate or deter workers to telecommute and how such factors contribute to their decision to continue telecommuting when the pandemic is over. To do so, we utilized the generalized structural equation modeling (GSEM) framework to better account for the interrelationship among the factors influencing the telecommuting behavior. The main source of data used in this study is a multidimensional activity-travel behavior survey which is recently conducted in the Chicago metropolitan area and comprises a rich set of information regarding participants’ socio-demographic details, health-related background, and information about the adjustments they made to their daily activity-travel routines due to the pandemic. The results revealed the significant effect of the perceived productivity of working from home on people’s intentions towards telecommuting in the future. Besides, we found the underlying effects of two latent factors on the perceived productivity of working from home: 1) the convenience of working from home, and 2) the underlying distractibility drivers.
DISCLAIMER: All information shared in the TRB Annual Meeting Online Program is subject to change without notice. Changes, if necessary, will be updated in the Online Program and this page is the final authority on schedule information.