A Review of Data Analytic Applications in Road Traffic Safety. Part 1: Descriptive and Predictive Modeling

Abstract

This part of the review aims to reduce the start-up burden of data collection and descriptive analytics for statistical modeling and route optimization of risk associated with motor vehicles. From a data-driven bibliometric analysis, we show that the literature is divided into two disparate research streams: (a) predictive or explanatory models that attempt to understand and quantify crash risk based on different driving conditions, and (b) optimization techniques that focus on minimizing crash risk through route/path-selection and rest-break scheduling. Translation of research outcomes between these two streams is limited. To overcome this issue, we present publicly available high-quality data sources (different study designs, outcome variables, and predictor variables) and descriptive analytic techniques (data summarization, visualization, and dimension reduction) that can be used to achieve safer-routing and provide code to facilitate data collection/exploration by practitioners/researchers. Then, we review the statistical and machine learning models used for crash risk modeling. We show that (near) real-time crash risk is rarely considered, which might explain why the optimization models (reviewed in Part 2) have not capitalized on the research outcomes from the first stream.

Keywords: crash risk modeling; data visualization; descriptive analytics; highway safety; predictive analytics.

Figure 1

A bibliographic analysis of the literature using the bibliometrix package in R.

Figure 2

A hierarchical view of outcome variables in crash risk modeling studies. The first level captures the data type, the second level shows the frequency, and the third level highlights examples and sources. * Acronyms: FMCSA = Federal Motor Carrier Safety Administration, NHTSA = National Highway Traffic Safety Administration, VT = Virginia Tech. ** Code: To simplify the data collection process, we present the R code needed to scrape and clean these different data sources at: https://caimiao0714.github.io/TrafficSafetyReviewRmarkdown/.

Figure 3

A hierarchy of predictor variables used in modeling crash risk. The first level captures the data type, the second level shows the frequency, and the third level highlights examples and sources. * Acronyms: AADT = Annual Average Daily Traffic, FHWA = U.S. Federal Highway Administration, DoT = U.S. Department of Transportation, and NOAA = U.S. National Oceanic & Atmospheric Administration. ** Code: To simplify the data collection process, we present the R code needed to scrape and clean these different data sources at: https://caimiao0714.github.io/TrafficSafetyReviewRmarkdown/.

Figure 4

Exploratory data analysis (EDA) goals and their associated techniques/methodological frameworks.

Figure 5

Symbol map showing the location of vehicle occupants killed in speed-related crashes in the US in December, 2016. The dashboard is available at [40].