|کد مقاله||کد نشریه||سال انتشار||مقاله انگلیسی||ترجمه فارسی||نسخه تمام متن|
|108430||161919||2013||7 صفحه PDF||سفارش دهید||دانلود رایگان|
In order to monitor the evolution of traffic on main city roads, a fixed-point data monitoring system is devised. Firstly, after qualitative traffic data (flow, speed, and occupancy) are acquired, they are then translated into the traffic qualitative state (congested or uncongested) by Fuzzy C-means Clustering algorithm. Secondly, the Congestion Evolution Index is determined using Rescaled Range Analysis of data mining. Finally, by taking a sequence pattern similarity measurement and applying condensed hierarchical clustering methods, the routine pattern is distinguished. Consequently, real-time outlier detection is realized by a distance-based outlier detection algorithm. This algorithm was successfully applied based on 11 days of fixed-point data on the eastern segment of the Shanghai North-South expressway, it is concluded that the outliers distributed in 12:10-13:20, 13:40-14:30 and 17:10-17:15 on September 30.
Journal: Journal of Transportation Systems Engineering and Information Technology - Volume 13, Issue 5, October 2013, Pages 30–36