详情

原文传递 Identifying and Classifying Highway Bottlenecks Based on Spatial and Temporal Variation of Speed

题名：	Identifying and Classifying Highway Bottlenecks Based on Spatial and Temporal Variation of Speed
其他题名：	Ahlawat,K.,and A.P.Singh.2017."A novel hybrid technique for big data classification using decision tree learning."In Proc.,Int.Conf.on Computational Intelligence,Communications,and Business Analytics,118-128.Singapore:Springer.
正文语种：	英文
作者：	Roshan Jose
关键词：	bottleneck；identification；structure；presence；prediction；management；variations；class；lassi；profile
摘要：	In this paper, we develop a novel approach to identify bottlenecks on highways using probe data collected by commercial global positioning system (GPS) fleet management devices installed in trucks. Further, the bottlenecks are classified based on the type of infrastructure present. Three main tasks were undertaken: (1) identification and classification of infrastructure at highway bottleneck locations, (2) examination and comparison between the various types of bottlenecks, and finally (3) prediction and classification of bottlenecks on highways using a decision tree classifier. Spatial and temporal variations of speed profile were primarily used for the identification and classification of bottlenecks. The results show that different types of bottlenecks due to construction zones, the presence of intersections, and toll plazas can be identified with high accuracy. Additionally, the presence of flyovers and bridges can also be detected from this speed profile. The findings of this study show that GPS data can not only be used to predict the locations of bottlenecks but also provide insightful information about the type of infrastructure, which is useful for highway operations and management.
出版年：	2018
论文唯一标识：	P-72Y2018V144N12004
英文栏目名称：	TECHNICAL PAPERS
doi：	10.1061/JTEPBS.0000183
期刊名称：	Journal of Transportation Engineering
拼音刊名(出版物代码)：	P-72
卷：	144
期：	12
页码：	27-38