Study of Signalized Intersection Crashes Using Artificial Intelligence Methods

Integrating Before and After Crash Features Into Measuring the Effectiveness of Intersection Safety Improvement Project in Harbin / Prieš Ir Po Eismo Įvykio Atsiradusių Padarinių Visuma, Vertinant Saugaus Eismo Sankryžose Projekto Efektyvumą Charbino Mieste / Учет Совокупности Факторов, Возникающих До И После Дорожно-Транспортного Происшествия, Для Оценки Эффективности Проекта По Безопасному Движению На Перекрестках В Городе Харбине

“…relationships between factors and intersection involved crashes (Pernía et al 2002;Wang et al 2006;Murphy, Hummer 2007;Tay, Rifaat 2007;Liu et al 2008). Another class of studies utilizes time series techniques for identifying whether a change in a policy at a given time had an impact on accident rates, thus analyzing relationships with other predictors by comparing trends in predictors and fatal accidents and developing a statistical model for forecasting future fatality rates such as in research works of Bared et al (2005), Kim et al (2007), Ye et al (2009), Abdel-Aty et al (2009), Mitra (2009), Al-Harthei et al (2010), etc.…”

A Framework for Traffic Collision Prediction Using Historical Accident Information and Real-Time Sensor Data: A Case Study for the City of Ottawa

“…Various forms have been employed in the literature for traffic accidents: backpropagation networks in [26], [27], probabilistic neural networks in [28], [29], and wavelet neural networks in [30]. A form of neural network, support vector machines (SVM) has also been used in the context of traffic accidents in several publications [31], [32], [33], [34].…”

“…A backpropagation neural network [27] is trained over data representing 102 signalized intersections and 3441 crash records from 1999-2004. The network has 34 inputs representing the road width, the percentage of motorcycles, the complexity of signal timing scenario, the overall traffic volume, the percentage of left turning vehicles, the signal cycle time, the directional traffic volume, the existence of warning signs, the median type between fast and slow traffic lanes, the obstacles within traffic lanes, and the percentage of right turning vehicles and predicts the number of accidents.…”

“…Another popular solution for traffic data analysis and for the identification of accident causes is decision trees [47], [46], [27], [48], [32], [49], [50], [51], [52], [53].…”

“…Joint probability regression is used in [54] to model the frequency and severity intersections and 3441 crashes in [27]. With these factors, a decision tree is applied to categorize intersections' safety level (safe, fairly safe, fairly unsafe, fairly dangerous, dangerous, extremely dangerous).…”