Traffic Congestion Index and Level Estimation using Two Phase Fuzzy Controller

Author Name(s): Prof. Prachiti Pimple, Prof. Gajanan Babhulkar, Prof. Bhumesh Masram, Prof. Payal Jadhav, Prof. Suraj Bandichode
Author Email: pspimple@kkwagh.edu.in

Abstract

Many measures have been proposed to represent the status of traffic conditions on arterial roadways in urban areas. Traffic congestion is rising nowadays and to understand its nature, a systematic mechanism is required. A new approach is presented in this research work to measure the congestion index first and then the congestion level. In this research work, a two-phase fuzzy controller is applied wherein in first phase the traffic congestion index is measured by using travel speed rate and very-low speed rate followed by congestion level measurement by using density state and congestion index in next phase. The application of the proposed approach is demonstrated using real-world data of small area segment of Nagpur city, India. The outcome was a single congestion index value between 0 and 1, where 0 is the best condition and 1 is the worst condition.

Keywords

Traffic Congestion Level Estimation, Congestion Index, Travel Speed Rate, Very-Low Speed Rate

Introduction

Traffic congestion is rising nowadays and to understand its nature, a systematic mechanism is required. A new approach is presented in this research work to measure the congestion index first and then the congestion level. A two-phase fuzzy controller is applied wherein in first phase the traffic congestion index is measured by using travel speed rate and very-low speed rate followed by congestion level measurement by using density state and congestion index in next phase. The application of the proposed approach is demonstrated using real-world data of small area segment of Nagpur city, India. Problems need to be considered while measuring congestion are described next, Measurements and observations are approximate. Imprecision exists in both demand-side and supply-side. Intersection signal timings and geometric design parameters from supply-side, varying delay and travel time from demand-side are not precise and hence accuracy in measuring congestion is associated with the uncertainty.

Acceptable speed condition changes depending on traveler’s experience and roadway condition; if the path segment geometry is good then travelers feel that acceptable speed is high and if the path segment geometry is poor then travelers feel that acceptable speed is low.

A use of Level-of-Service (LOS) is misleading sometimes when the condition is nearer to threshold. A small change in the input makes significance difference in output.

Conclusion

The proposed approach considers the travel speed rate and the vary-low speed rate and produces a composite measure as Congestion Index. The fuzzy inference system is applied because possible errors in computation or measurement cause the individual values to be imprecise and the implication of the values with respect to the severity of congestion is ambiguous. The inference process is based on natural-language rules, which are consistent with the general feelings of the travellers. The proposed fuzzy controller is applied on real world datasets of Nagpur City, India for 31 signalized intersections and results are compared with highway capacity manual.

320 total views, 6 views today

Download PDF File

About the author: admin