ABSTRACT
With the advance of intelligent transportation system technologies, contributing factors to crashes can be obtained in real time. Analyzing these factors can be critical in improving traffic safety. Despite many crash models having been successfully developed for safety analytics, most models associate crash observations and contributing factors at the aggregate level, resulting in potential information loss. This study proposes an efficient Gaussian process modulated renewal process model for safety analytics that does not suffer from information loss due to data aggregations. The proposed model can infer crash intensities in the continuous-time dimension so that they can be better associated with contributing factors that change over time. Moreover, the model can infer non-homogeneous intensities by relaxing the independent and identically distributed (i.i.d.) exponential assumption of the crash intervals. To demonstrate the validity and advantages of this proposed model, an empirical study examining the impacts of the COVID-19 pandemic on traffic safety at six interstate highway sections is performed. The accuracy of our proposed renewal model is verified by comparing the areas under the curve (AUC) of the inferred crash intensity function with the actual crash counts. Residual box plot shows that our proposed models have lower biases and variances compared with Poisson and Negative binomial models. Counterfactual crash intensities are then predicted conditioned on exogenous variables at the crash time. Time-varying safety impacts such as bimodal, unimodal, and parabolic patterns are observed at the selected highways. The case study shows the proposed model enables safety analytics at a granular level and provides a more detailed insight into the time-varying safety risk in a changing environment.
Subject(s)
Automobile Driving , COVID-19 , Accidents, Traffic/prevention & control , Humans , Models, Statistical , Pandemics , SafetyABSTRACT
On March 22, 2020, the State of New York issued a “stay-at-home” policy, wherein all non-essential businesses were on pause until June 8, 2020. The bike-sharing system (BSS) and yellow taxi system (YTS) in Manhattan were substantially affected. This sudden drop in demand can impact not only short and long-term mobility but also the sustainability of transport network. Given that few empirical studies are focusing on the impacts of the “stay-at-home” policy on the BSS and YTS, this further substantiates the importance of analyzing how the policy affects the overall transportation system in New York City (NYC). This paper aims to fill this gap by quantifying the impacts of the “stay-at-home” policy on the two aforementioned transportation systems. Specifically, the following three research gaps are summarized in this study: I) The hidden biases in current “stay-at-home” policy estimation methods were not properly addressed;II) The policy impacts on BSS and YTS during different periods of the effective day were unclear;III) The sensitivity of uncontrolled confounders in long-term policy impact estimations was poorly discussed. We addressed these important research gaps by introducing robust statistical approaches like regression discontinuity design (RDD) and propensity score matching (PSM) methods, which can overcome methodological challenges such as counterfactual restoration, spatiotemporal heterogeneities, and unmeasured confounders. The BSS and YTS were studied at the aggregated neighborhood levels. Results demonstrate that the impacts to BSS have higher variations than YTS usage. The monthly average treatment effects on the treated (ATT) for BSS ranged from -72% to -28% respectively in March and June, while YTS ranged from -96% to -94%. Evidence suggests that demand for BSS surged on weekends in May and June. Understanding the impact of this short-term yet significant policy change on travel behavior will help optimize supply and demand management strategies, thereby improving the long-term sustainability should similar situations arise in the future.