The potential stickiness of pandemic-induced behavior changes in the United States.

Human behavior is notoriously difficult to change, but a disruption of the magnitude of the COVID-19 pandemic has the potential to bring about long-term behavioral changes. During the pandemic, people have been forced to experience new ways of interacting, working, learning, shopping, traveling, and eating meals. A critical question going forward is how these experiences have actually changed preferences and habits in ways that might persist after the pandemic ends. Many observers have suggested theories about what the future will bring, but concrete evidence has been lacking. We present evidence on how much US adults expect their own postpandemic choices to differ from their prepandemic lifestyles in the areas of telecommuting, restaurant patronage, air travel, online shopping, transit use, car commuting, uptake of walking and biking, and home location. The analysis is based on a nationally representative survey dataset collected between July and October 2020. Key findings include that the "new normal" will feature a doubling of telecommuting, reduced air travel, and improved quality of life for some.

The COVID-19 pandemic and the future of telecommuting in the United States.

This study focuses on an important transport-related long-term effect of the COVID-19 pandemic in the United States: an increase in telecommuting. Analyzing a nationally representative panel survey of adults, we find that 40-50% of workers expect to telecommute at least a few times per month post-pandemic, up from 24% pre-COVID. If given the option, 90-95% of those who first telecommuted during the pandemic plan to continue the practice regularly. We also find that new telecommuters are demographically similar to pre-COVID telecommuters. Both pre- and post-COVID, higher educational attainment and income, together with certain job categories, largely determine whether workers have the option to telecommute. Despite growth in telecommuting, approximately half of workers expect to remain unable to telecommute and between 2/3 and 3/4 of workers expect their post-pandemic telecommuting patterns to be unchanged from their pre-COVID patterns. This limits the contribution telecommuting can make to reducing peak hour transport demand.

Interrelationships between urban travel demand and electricity consumption: a deep learning approach.

The analysis of infrastructure use data in relation to other components of the infrastructure can help better understand the interrelationships between infrastructures to eventually enhance their sustainability and resilience. In this study, we focus on electricity consumption and travel demand. In short, the premise is that when people are in buildings consuming electricity, they are not generating traffic on roads, and vice versa, hence the presence of interrelationships. We use Long Short Term Memory (LSTM) networks to model electricity consumption patterns of zip codes based on the traffic volume of the same zip code and nearby zip codes. For this, we merge two datasets for November 2017 in Chicago: (1) aggregated electricity use data in 30-min intervals within the city of Chicago and (2) traffic volume data captured on the Chicago expressway network. Four analyses are conducted to identify interrelationships: (a) correlation between two time series, (b) temporal relationships, (c) spatial relationships, and (d) prediction of electricity consumption based on the total traffic volume. Overall, from over 250 models, we identify and discuss complex interrelationships between travel demand and electricity consumption. We also analyze and discuss how and why model performance varies across Chicago.

A database of travel-related behaviors and attitudes before, during, and after COVID-19 in the United States.

The COVID-19 pandemic has impacted billions of people around the world. To capture some of these impacts in the United States, we are conducting a nationwide longitudinal survey collecting information about activity and travel-related behaviors and attitudes before, during, and after the COVID-19 pandemic. The survey questions cover a wide range of topics including commuting, daily travel, air travel, working from home, online learning, shopping, and risk perception, along with attitudinal, socioeconomic, and demographic information. The survey is deployed over multiple waves to the same respondents to monitor how behaviors and attitudes evolve over time. Version 1.0 of the survey contains 8,723 responses that are publicly available. This article details the methodology adopted for the collection, cleaning, and processing of the data. In addition, the data are weighted to be representative of national and regional demographics. This survey dataset can aid researchers, policymakers, businesses, and government agencies in understanding both the extent of behavioral shifts and the likelihood that changes in behaviors will persist after COVID-19.

How is COVID-19 reshaping activity-travel behavior? Evidence from a comprehensive survey in Chicago.

The novel COVID-19 pandemic has caused upheaval around the world and has led to drastic changes in our daily routines. Long-established routines such as commuting to workplace and in-store shopping are being replaced by telecommuting and online shopping. Many of these shifts were already underway for a long time, but the pandemic has accelerated them remarkably. This research is an effort to investigate how and to what extent people's mobility-styles and habitual travel behaviors have changed during the COVID-19 pandemic and to explore whether these changes will persist afterward or will bounce back to the pre-pandemic situation. To do so, a stated preference-revealed preference (SP-RP) survey is designed and implemented in the Chicago metropolitan area. The survey incorporates a comprehensive set of questions associated with individuals' travel behaviors, habits, and perceptions before and during the pandemic, as well as their expectations about the future. Analysis of the collected data reveals significant changes in various aspects of people's travel behavior. We also provide several insights for policymakers to be able to proactively plan for more equitable, sustainable, and resilient cities.

Investigating the injury severity of single-vehicle truck crashes in a developing country.

Trucking plays a vital role in economic development in every country, especially countries where it serves as the backbone of the economy. The fast growth of economy in Iran as a developing country has also been accompanied by an alarming situation in terms of fatalities in truck-involved crashes, among the drivers and passengers of the trucks as well as the other vehicles involved. Despite the sizable efforts to investigate the truck-involved crashes, very little is known about the safety of truck movements in developing countries, and about the single-truck crashes worldwide. Thus, this study aims to uncover significant factors associated with injury severities sustained by truck drivers in single-vehicle truck crashes in Iran. The explanatory factors tested in the models include the characteristics of drivers, vehicles, and roadways. A random threshold random parameters hierarchical ordered probit model is utilized to consider heterogeneity across observations. Several variables turned out to be significant in the model, including driver's education, advanced braking system deployment, presence of curves on roadways, and high speed-limit. Using those results, we propose safety countermeasures in three categories of 1) educational, 2) technological, and 3) road engineering to mitigate the severity of single-vehicle truck crashes.

Toward safer highways, application of XGBoost and SHAP for real-time accident detection and feature analysis.

Detecting traffic accidents as rapidly as possible is essential for traffic safety. In this study, we use eXtreme Gradient Boosting (XGBoost)-a Machine Learning (ML) technique-to detect the occurrence of accidents using a set of real time data comprised of traffic, network, demographic, land use, and weather features. The data used from the Chicago metropolitan expressways was collected between December 2016 and December 2017, and it includes 244 traffic accidents and 6073 non-accident cases. In addition, SHAP (SHapley Additive exPlanation) is employed to interpret the results and analyze the importance of individual features. The results show that XGBoost can detect accidents robustly with an accuracy, detection rate, and a false alarm rate of 99 %, 79 %, and 0.16 %, respectively. Several traffic related features, especially difference of speed between 5 min before and 5 min after an accident, are found to have relatively more impact on the occurrence of accidents. Furthermore, a feature dependency analysis is conducted for three pairs of features. First, average daily traffic and speed after accidents/non-accidents time at the upstream location are interpreted jointly. Then, distance to Central Business District and residential density are analyzed. Finally, speed after accidents/non-accidents time at upstream location and speed after accidents/non-accidents time at downstream location are evaluated with respect to the model's output.

Real-time accident detection: Coping with imbalanced data.

Detecting accidents is of great importance since they often impose significant delay and inconvenience to road users. This study compares the performance of two popular machine learning models, Support Vector Machine (SVM) and Probabilistic Neural Network (PNN), to detect the occurrence of accidents on the Eisenhower expressway in Chicago. Accordingly, since the detection of accidents should be as rapid as possible, seven models are trained and tested for each machine learning technique, using traffic condition data from 1 to 7 min after the actual occurrence. The main sources of data used in this study consist of weather condition, accident, and loop detector data. Furthermore, to overcome the problem of imbalanced data (i.e., underrepresentation of accidents in the dataset), the Synthetic Minority Oversampling TEchnique (SMOTE) is used. The results show that although SVM achieves overall higher accuracy, PNN outperforms SVM regarding the Detection Rate (DR) (i.e., percentage of correct accident detections). In addition, while both models perform best at 5 min after the occurrence of accidents, models trained at 3 or 4 min after the occurrence of an accident detect accidents more rapidly while performing reasonably well. Lastly, a sensitivity analysis of PNN for Time-To-Detection (TTD) reveals that the speed difference between upstream and downstream of accidents location is particularly significant to detect the occurrence of accidents.