Non-parametric machine learning methods for evaluating the effects of traffic accident duration on freeways
Journal Title: Archives of Transport - Year 2017, Vol 43, Issue 3
Abstract
Traffic accidents usually cause congestion and increase travel-times. The cost of extra travel time and fuel consumption due to congestion is huge. Traffic operators and drivers expect an accurately forecasted accident duration to reduce uncertainty and to enable the implementation of appropriate strategies. This study demonstrates two non-parametric machine learning methods, namely the k-nearest neighbour method and artificial neural network method, to construct accident duration prediction models. The factors influencing the occurrence of accidents are numerous and complex. To capture this phenomenon and improve the performance of accident duration prediction, the models incorporated various data including accident characteristics, traffic data, illumination, weather conditions, and road geometry characteristics. All raw data are collected from two public agencies and were integrated and cross-checked. Before model development, a correlation analysis was performed to reduce the scale of interrelated features or variables. Based on the performance comparison results, an artificial neural network model can provide good and reasonable prediction for accident duration with mean absolute percentage error values less than 30%, which are better than the prediction results of a k-nearest neighbour model. Based on comparison results for circumstances, the Model which incorporated significant variables and employed the ANN method can provide a more accurate prediction of accident duration when the circumstances involved the day time or drunk driving than those that involved night time and did not involve drunk driving. Empirical evaluation results reveal that significant variables possess a major influence on accident duration prediction.<br/><br/>
Authors and Affiliations
Ying Lee, Chien-Hung Wei, Kai-Chon Chao
Keywords
Related Articles
- EP ID EP220395
- DOI 10.5604/01.3001.0010.4228
- Views 105
- Downloads 0
Using Queue Theory to Analysis and Evaluation of the Logistics Centre Workload
The article presents an application of queuing theory (mass service) to describe transportation process. It presents the essential concepts of the queuing theory. The process of moving a load unit through the logistics c...
Designing and efficiency of database for simulation of processes in systems. Case study for the simulation of warehouse processes
The basis of each IT system is data exchange both between individual modules of program and between program and user. The number of exchanged data and the number of operations on the data, which should be performed, requ...
Evaluation of influence of mobility management instruments implemented in separated areas of the city on the changes in modal split
The article presents results of research aimed at construction of the model for evaluation of potential impact of mobility management instruments implemented in areas of high volume of work related trips. The model helps...
The Influence of Damping Changes on Vertical Dynamic Loads of Wheel – Experimental Investigations
In this paper the results of tests of vertical dynamic wheel loads due to changes in suspension damping level are presented. The changes in damping level were performed using semi-active shock absorbers with by-pass valv...
Dual ORC-Brayton power system for waste heat recovery in heavy-duty vehicles
Reducing the amount of energy required in industrial activities is one of the proven ways to achieve major cost savings, especially in the face of soaring energy prices. In the transport sector, besides the financial ben...