Assessing Safety on Dutch Freeways with Data from Infrastructure-Based Intelligent Transportation Systems

Abstract

Most freeway traffic surveillance technologies deployed around the world remain infrastructure based, with underground loop detectors being the most common among them. A proactive application for traffic surveillance data recently explored for some freeways in the United States is the estimation of real-time crash risk. The application involves establishing relationships between historical crashes and archived traffic data collected before those crashes. In these studies, crash occurrence on freeway sections has been related to temporal-spatial variation in speed and high lane occupancy. Critical modeling questions that remain unanswered relate to transferability of such an approach. This study attempts to address the issues of such transfer through analysis of crash data and corresponding loop detector data from five freeways in the Utrecht region of the Netherlands. Traffic surveillance systems for these freeways include more detectors per kilometer than most U.S. freeways. Their real-time data are also already being used for applications of advanced intelligent transportation systems. The analysis procedure proposed here accounts for these distinctions. In addition to these transferability issues, application is introduced of a new data-mining methodology, Random Forests, for identifying variables significantly associated with the binary target variable (crash versus noncrash). It was found that the average and standard deviations of speed and volume are related to real-time crash likelihood. Subjecting these significantly related variables to multilayer perceptron and normal radial basis function neural networks resulted in classifiers that achieved classification accuracy of approximately 61% for crashes and 79% for noncrashes. The promising classification accuracy indicates that these models can be used for reliable assessment of real-time crash risk on Dutch freeways as well.

Big Data applications in real-time traffic operation and safety monitoring and improvement on urban expressways

“…Statistical methods such as simple/matched-case logistic regression (Abdel-Aty et al, 2004) and Bayesian statistics (Abdel-Aty et al, 2012) present the effects of candidate variables in a more interpretable way. Data mining based methods could be neural networks (Pande and Abdel-Aty, 2006), random forests (Abdel-Aty et al, 2008), and support vector machines , etc. Data mining methods are applauded for their high prediction accuracy but criticized for the black-box-like process.…”

A Bibliometric Analysis of the Intelligent Transportation Systems Research Based on Science Mapping

“…In the next period the thematic area developed and it was divided in two main topics: GPS [40] and vehicle routing in real time [41]. Finally, in the period 2007-2011, the thematic area focused on the topic GPS, as well as in improving its accuracy, availability, and continuity of service [42]. The presented analysis provides a complete view of the conceptual structure of the ITS research field, giving us the ability to uncover the different topics researched by the ITS community from 1992 to 2011.…”

A novel variable selection method based on frequent pattern tree for real-time traffic accident risk prediction

“…Most of the previous studies considered more than one detector during the extraction process, such as one upstream detector and one downstream detector (Abdel-Aty et al, 2008), and two upstream detectors, two downstream detectors and one detector covering the accident location (Hossain and Muromachi, 2012). In this paper, due to the problem of missing data, we were forced to rely on only one detector, that is to say, the one reporting an accident.…”

“…As a classification problem, the pre-crash condition and normal traffic condition have to be defined first (Hossain and Muromachi, 2012). Some studies defined the pre-crash condition as a time period starting right before an accident and extending up to 5 or 10 min (Oh et al, 2005;Zheng et al, 2010), while some studies defined it as a 5 min time period starting from a close time point such as 4 or 5 min before the accident (Abdel-Aty et al, 2008;Hossain and Muromachi, 2012). In this paper, as shown in Fig.…”