Exploiting Surrogate Safety Measures and Road Design Characteristics towards Crash Investigations in Motorway Segments

Abstract

High quality data on road crashes, road design characteristics, and traffic are typically required to predict crash frequency. Surrogate Safety Measures (SSMs) are an alternative category of indicators that can be used in road safety analyses in order to quantify various unsafe traffic events. The objective of this research is to exploit road geometry data and SSMs toward various road crash investigations in motorway segments. To that end, for this analysis, a database containing data on injury and property-damage-only crashes, road design characteristics, and SSMs of 668 segments was compiled and utilized. The results of the developed negative binomial regression model revealed that crash frequency is positively correlated with the average annual daily traffic volume, the length of the segment, harsh accelerations, and harsh braking. Moreover, four distinct clusters representing crash risk levels of the examined segments emerged from the hierarchical clustering procedure, ranging from more risk-prone, potentially unsafe locations to more safe locations. These four clusters also formed the response variable classes of a random forest model. This classification model used various road geometry data and SSMs as predictors and achieved high classification performance for all classes, averaging more than 88% correct classification rates.

Comparing Machine Learning Techniques for Predictions of Motorway Segment Crash Risk Level

“…It can be observed that this representative motorway segment is more likely to belong to the lowest crash risk level, as denoted by the positive (green) bars of all predictors for this specific class. This crash risk level corresponds to overall safer locations with lower traffic volumes and road crashes by segment length than the motorway segments between the first and the third crash risk level [17]. It is worth noting that Figure 2 shows the contribution of only a subset of the variables that have been included in the multiclass classification RF model, as the other variables are not contributing much to the model's predictions and their contribution to the model's output can be, therefore, considered negligible.…”

“…In addition, the Shapley additive explanations (SHAP) were estimated and provided as an attempt to deal with the interpretation of the best performing machine learning algorithm outcomes. It should be noted that the present research is an extension of the study that was conducted by Nikolaou et al, in which the motorway segments' risk levels were determined and the comparison of various ML classification models was proposed as a future research direction [17].…”

“…The length of the considered segments varies from 200 m to 600 m. It is noted that tunnel and toll station segments were not included in the analysis. For further explanations on the hierarchical clustering process and the extraction of both road design characteristics and naturalistic driving behaviour data the reader is referred to Nikolaou et al [17]. The variables that were considered in the analyses of this research are presented in Table 1.…”

Investigating effects of future path visualisation on path choices during collision encounters

“…Following the definition proposed in Parker et al, 1988, a traffic conflict is "an event involving two or more road users, in which the action of one user causes the other user to make an evasive maneuver to avoid a collision". Such motion conflicts are commonly evaluated using surrogate safety measures (Nikolaou et al, 2023). While other methods to characterise a conflict have been proposed, the post encroachment time (PET) and time-to-collision (TTC) are commonly used surrogate safety measures to study conflict severity (Sayed et al, 1994, Hyden, 1987, Hayward, 1972, Orsini et al, 2023, Paul and Ghosh, 2020.…”

Analysis of Road Infrastructure and Traffic Factors Influencing Crash Frequency: Insights from Generalised Poisson Models

“…Transportation researchers utilise various techniques to understand the crash phenomenon, such as crash investigation reports, video analysis, naturalistic driving studies, simulation studies, crash data statistical analysis, artificial neural networks, surrogate safety measures, and telematics data analysis [4][5][6][7]. Safety performance functions (SPFs)statistical models for crash prediction-have been the subject of many studies during the past few decades [8][9][10][11].…”