By Topic

Using k-means clustering to identify time-of-day break points for traffic signal timing plans

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Xiaodong Wang ; Dept. of Civil & Environ. Eng., Utah Univ., Salt Lake City, UT, USA ; W. Cottrell ; Sichun Mu

The k-means method, a nonhierarchical clustering algorithm, is applied to traffic volume data to determine time-of-day (TOD) breakpoints for traffic signal timing plans. Other methods, including hierarchical clustering techniques, have been applied to traffic signal timings, but they are computationally intensive and require a substantial amount of data storage space. The procedure requires that the analyst specify the number of clusters prior to engaging the algorithm. The resultant allocations of volumes to clusters may be "noisy"; smoothing may be needed to avoid having an inoperable number of TOD breakpoints. The algorithm is applied to a small case study involving a two-intersection corridor and just under four hours of volume data. Three time intervals were identified, with a peak, two-hour period sandwiched by two off-peak segments. An expanded application of the algorithm on a longer corridor or network, over a longer time period, is recommended. Subsequent steps would be to develop the signal timing plans for the study intersections, evaluate the proposed plans, and assess the potential for their implementation. The k-means method can develop TOD breakpoints from traffic volumes, making it a potentially useful procedure where detectors supplying additional traffic information are either sparse or nonexistent.

Published in:

Proceedings. 2005 IEEE Intelligent Transportation Systems, 2005.

Date of Conference:

13-15 Sept. 2005