An Improved Deep Learning Model for Traffic Crash Prediction

Abstract

Machine-learning technology powers many aspects of modern society. Compared to the conventional machine learning techniques that were limited in processing natural data in the raw form, deep learning allows computational models to learn representations of data with multiple levels of abstraction. In this study, an improved deep learning model is proposed to explore the complex interactions among roadways, traffic, environmental elements, and traffic crashes. The proposed model includes two modules, an unsupervised feature learning module to identify functional network between the explanatory variables and the feature representations and a supervised fine tuning module to perform traffic crash prediction. To address the unobserved heterogeneity issues in the traffic crash prediction, a multivariate negative binomial (MVNB) model is embedding into the supervised fine tuning module as a regression layer. The proposed model was applied to the dataset that was collected from Knox County in Tennessee to validate the performances. The results indicate that the feature learning module identifies relational information between the explanatory variables and the feature representations, which reduces the dimensionality of the input and preserves the original information. The proposed model that includes the MVNB regression layer in the supervised fine tuning module can better account for differential distribution patterns in traffic crashes across injury severities and provides superior traffic crash predictions. The findings suggest that the proposed model is a superior alternative for traffic crash predictions and the average accuracy of the prediction that was measured by RMSD can be improved by 84.58% and 158.27% compared to the deep learning model without the regression layer and the SVM model, respectively.

Truck Traffic Flow Prediction Based on LSTM and GRU Methods With Sampled GPS Data

“…Var calculates the correlation coefficient. (6) It is worth noting that the sequence of expansion and prediction might affect prediction accuracy. Based on previous-expansion, post-prediction, the expanding and forecasting sequence of the model is reversed to obtain the truck traffic flow result of previous-prediction, postexpansion.…”

Application of deep learning techniques in predicting motorcycle crash severity

“…This network is named as feedforward because there are no feedback connections in which output of the model are fed back into itself. [20][21][22][23] Neural network was developed from single neuron named perceptron. It categorizes a set of input into one of two classes.…”

Civil Infrastructure Management Models for the Connected and Automated Vehicles Technology

“…Many studies for crashes at intersections have resulted in crash prediction models that could estimate the probability of crash occurrence. Dong et al [26] reported on a crash prediction model based on data from Knox County in Tennessee, U.S.A. From this study, it was observed that, for the selected 635 roadway segments, from 2010 to 2014, a total of 5365 traffic crashes reported included 135 (2.51%) major injury crashes (fatal and incapacitating injury), 1312 (24.46%) minor injury crashes, and 3917 (73.02%) PDO crashes. Table 2, shown above, can be further refined by applying an appropriate probability of occurrence to estimate the expected costs.…”

“…The CPI values were also used to adjust the expected cost values to the 2019 dollars (Table 3). Based on Dong et al [26], the severity probabilities were adopted on the previously defined KABCO scale as follows: Fatal Injury (K) (1.25%); Suspected Serious Injury (A) (1.26%); Suspected Minor Injury (B) (12.23%); Possible Injury (C) (12.23%); and No Apparent Injury (O) (73.02%). Using these crash severity proportions, the crash cost for category A can be estimated to be 0.0125 * $11,295,402, or $142,322.07, which we can round up to $150,000.…”