Traffic flow analysis and management are among the most effective ways to improving traffic and mitigating its unfortunate consequences. In the field of traffic engineering, traffic and its various aspects are defined by analyzing variables such as quantity, velocity, and density. This article addresses the challenge of appropriately dealing with the uncertainty of traffic variables and converting traffic data into understandable verbal expressions for drivers and urban planners. The study utilizes a clustering approach to analyzing traffic variables and determining the traffic condition. A new fuzzy clustering method has been developed to enhance the performance of clustering methods, which is then used to detect abnormal traffic conditions on a route based on the value of traffic variables. The algorithm and proposed method have been evaluated on the traffic dataset of a high-traffic route in Tehran, the capital of Iran. The implementation results demonstrate the traffic conditions on the selected route can be shown in six clusters or states.
Clustering Approach to Identifying and Analyzing the Traffic Conditions: A Novel Hybrid Cloud Density and Fuzzy Clustering Algorithm
Int. J. ITS Res.
2025-04-01
15 pages
Article (Journal)
Electronic Resource
English
Traffic analysis , Intelligent transportation systems , Traffic variables , Cloud density , Fuzzy clustering Engineering , Electrical Engineering , Automotive Engineering , Robotics and Automation , Computer Imaging, Vision, Pattern Recognition and Graphics , Civil Engineering , User Interfaces and Human Computer Interaction
An Extreme Value Based Neural Clustering Approach for Identifying Traffic States
British Library Conference Proceedings | 2005
|