Estimation of Traffic Incident Duration: A Comparative Study of Decision Tree Models
- PDF / 960,814 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 62 Downloads / 153 Views
RESEARCH ARTICLE-CIVIL ENGINEERING
Estimation of Traffic Incident Duration: A Comparative Study of Decision Tree Models Abdulsamet Saracoglu1 · Halit Ozen1 Received: 23 January 2020 / Accepted: 30 April 2020 © King Fahd University of Petroleum & Minerals 2020
Abstract Unexpected events such as crashes, disabled vehicles, flat tires and spilled loads cause traffic congestion or extend the duration of the traffic congestion on the roadways. It is possible to reduce the effects of such incidents by implementing intelligent transportation systems solutions that require the estimation of the incident duration to identify well-fitted strategies. This paper presents a methodology to establish incident duration estimation models by utilizing decision tree models of CHAID, CART, C4.5 and LMT. For this study, the data contained traffic incidents that occurred on the Istanbul Trans European Motorway were obtained and separated into three groups according to duration by utilizing some studies about classification of traffic incidents. By using classified data, decision tree models of CHAID, CART, C4.5 and LMT were established and validated to estimate the incident duration. According to the results, although the models used different variables, the decision tree models of CHAID, CART and C4.5 have nearly the same prediction accuracy which is approximately 74%. On the other hand, the prediction accuracy of decision tree model of LMT is 75.4% which is somewhat better than the others. However, C4.5 model required less number of parameters than the others, while its accuracy is the same with others. Keywords CART · CHAID · C4.5 · Decision tree · LMT · Traffic incident duration
1 Introduction Traffic congestion is one of the major problems met in the operation of highways in large urban areas. It often occurs in the road parts where demand is greater than capacity and emerges in two different ways including recurring and nonrecurring congestion. Travelers can plan their trips according to the expected congestion in the recurring congestion, which occurs on the peak hour of traffic every regular weekday. However, non-recurring congestion originated by traffic incidents, such as crashes, vehicle breakdowns and debris, causes unexpected congestion on the roadway. Unless they are informed about the incident before starting their trips, travelers cannot estimate this congestion and do not change their trip planning. So, this causes an increase in congestion and in the probability of secondary incidents occurring, etc. To minimize the negative effects of traffic incidents,
* Abdulsamet Saracoglu [email protected] 1
Department of Civil Engineering, Yildiz Technical University, Istanbul, Turkey
advanced Traffic Incident Management (TIM) systems have to be developed and applied. The advanced TIM systems primarily require to analyze the incident process in detail and predict incident duration based on environmental conditions and incident characteristics (incident type, part of day, incident severity, weather condition, etc.
Data Loading...
