Crash Density and Severity Prediction Using Recurrent Neural Networks Combined with Particle Swarm Optimization

Predicting traffic crashes has been an important topic of traffic safety research for the past many years. This paper investigates the data from police crash reports provided by the Washington State Department of Transportation. The data consists of records of four years from January 2011 to December 2014 for three main interstate highways (including I-5, I-90, and I-405). A deep learning model using a recurrent neural network (RNN) combined with particle swarm optimization (PSO) is developed and employed to predict the crash density in different severity levels such as property damage only (PDO) and fatal-injury crashes, based on 48,154 crash records that have occurred. All the crash records are randomly divided into training set, validation set, and test set with the proportion ratio of 70, 15, and \(15\%\). The cross-validation is employed to prevent the model from over fit during the training period. A normalized probability-based PSO is designed for optimizing the identified significant factors which can improve the prediction accuracy. The weighted mean squared error (MSE) of the prediction result is employed to measure the performance of the developed model. Nine explanatory variables are selected from fifteen contributing factors. The proposed model is compared with generalized nonlinear model-based mixed multinomial logit approach (GNM-based mixed MNL). The results show that the new model has lower fatal-injury and PDO MSEs. Sensitivity analysis on the selected variables demonstrates the capability of the new model for generating interpretable parameters. The findings of this study provide new insights into the prediction of crash density and severity from the perspective of using roadway segment-based crash records.

[1]  Wei Fan,et al.  Identifying and Quantifying Factors Affecting Vehicle Crash Severity at Highway-Rail Grade Crossings: Models and Their Comparison , 2016 .

[2]  Jiuping Xu,et al.  Antithetic Method‐Based Particle Swarm Optimization for a Queuing Network Problem with Fuzzy Data in Concrete Transportation Systems , 2014, Comput. Aided Civ. Infrastructure Eng..

[3]  Mohamed Abdel-Aty,et al.  Development of Artificial Neural Network Models to Predict Driver Injury Severity in Traffic Accidents at Signalized Intersections , 2001 .

[4]  Jonathan E. Fieldsend,et al.  Using unconstrained elite archives for multiobjective optimization , 2003, IEEE Trans. Evol. Comput..

[5]  Mohamed M Ahmed,et al.  Effects of truck traffic on crash injury severity on rural highways in Wyoming using Bayesian binary logit models. , 2018, Accident; analysis and prevention.

[6]  Ren Gang,et al.  Traffic safety forecasting method by particle swarm optimization and support vector machine , 2011, Expert Syst. Appl..

[7]  Ramesh Sharda,et al.  Identifying significant predictors of injury severity in traffic accidents using a series of artificial neural networks. , 2006, Accident; analysis and prevention.

[8]  Fred L. Mannering,et al.  Analysis of vehicle accident-injury severities: A comparison of segment- versus accident-based latent class ordered probit models with class-probability functions , 2018, Analytic Methods in Accident Research.

[9]  Juan de Oña,et al.  Analysis of traffic accident severity using Decision Rules via Decision Trees , 2013, Expert Syst. Appl..

[10]  Jun Yan,et al.  Detecting traffic accident clusters with network kernel density estimation and local spatial statistics: an integrated approach , 2013 .

[11]  Denver Tolliver,et al.  Decision Tree Approach to Accident Prediction for Highway–Rail Grade Crossings: Empirical Analysis , 2016 .

[12]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[13]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[14]  Jürgen Teich,et al.  Strategies for finding good local guides in multi-objective particle swarm optimization (MOPSO) , 2003, Proceedings of the 2003 IEEE Swarm Intelligence Symposium. SIS'03 (Cat. No.03EX706).

[15]  Mohamed Abdel-Aty,et al.  Analysis of crash proportion by vehicle type at traffic analysis zone level: A mixed fractional split multinomial logit modeling approach with spatial effects. , 2018, Accident; analysis and prevention.

[16]  Mahdi Pour-Rouholamin,et al.  Wrong-way driving crashes: A random-parameters ordered probit analysis of injury severity. , 2018, Accident; analysis and prevention.

[17]  Madhar Taamneh,et al.  Clustering-based classification of road traffic accidents using hierarchical clustering and artificial neural networks , 2017, International journal of injury control and safety promotion.

[18]  Wei Fan,et al.  Key factors contributing to crash severity at highway-rail grade crossings , 2016 .

[19]  Helai Huang,et al.  A stable and optimized neural network model for crash injury severity prediction. , 2014, Accident; analysis and prevention.

[20]  Junjie Yang,et al.  A novel strategy of pareto-optimal solution searching in multi-objective particle swarm optimization (MOPSO) , 2009, Comput. Math. Appl..

[21]  Srinivas Reddy Geedipally,et al.  Analysis of crash severities using nested logit model--accounting for the underreporting of crashes. , 2012, Accident; analysis and prevention.

[22]  Wenhao Huang,et al.  Deep Architecture for Traffic Flow Prediction: Deep Belief Networks With Multitask Learning , 2014, IEEE Transactions on Intelligent Transportation Systems.

[23]  Amirfarrokh Iranitalab,et al.  Comparison of four statistical and machine learning methods for crash severity prediction. , 2017, Accident; analysis and prevention.

[24]  Mark Stevenson,et al.  Identifying behavioural change among drivers using Long Short-Term Memory recurrent neural networks , 2018 .

[25]  Azim Eskandarian,et al.  Vehicle crash modelling using recurrent neural networks , 1998 .

[26]  Naveen Eluru,et al.  Evaluating alternate discrete outcome frameworks for modeling crash injury severity. , 2013, Accident; analysis and prevention.

[27]  Jiuping Xu,et al.  A generalized nonlinear model-based mixed multinomial logit approach for crash data analysis. , 2017, Accident; analysis and prevention.

[28]  Biswajeet Pradhan,et al.  Severity Prediction of Traffic Accidents with Recurrent Neural Networks , 2017 .

[29]  Dominique Lord,et al.  Comparing Three Commonly Used Crash Severity Models on Sample Size Requirements: Multinomial Logit, Ordered Probit, and Mixed Logit Models , 2014 .