United States Department of Transportation
i
Incorporation of Speed/Travel-Time Data Sets in Traffic Performance Analysis
-
2019-09-01
-
[PDF-7.68 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final report, Jan 2016 - Jan 2019
-
Corporate Publisher:
-
Abstract:An accurate Travel Time (TT) information is essential for traffic prediction and analysis. Correct TT measurements can be used to reduce congestions, improve safety and enhance traffic flow. The work detailed in this paper reports on the development and implementation of an inexpensive Bluetooth traffic monitoring system for reliable and real-time TT measurements. The Bluetooth system proved instant and accurate TT measurements 99.9% of the time. Furthermore, algorithms used for collecting traffic information from various ODOT platforms were fully developed, integrated, and evaluated. National Performance Management Research Dataset (NPRMDS) was chosen for developing TT models, and for identifying TT outliers and anomalies. Since incidents on highways increase TT, artificial neural network (ANN) model was designed for incident classification based on many features, such as vehicle count, weather condition and traffic flow. The ANN model reported accuracy of 91%. Moreover, a Bayesian model for detecting non-recurrent congestions was developed and evaluated.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
