Surveillance trees: early detection of unusually high number of vehicle crashes

by Ross Sparks and Chris Okugami .

Abstract: The aim of the paper is to build a generalised multivariate outbreak detection methodology for Poisson counts data. The approach is applied to the problem of early detection of unusually high vehicle crashes. Multivariate clustered outbreaks are searched for in the dimensions of age and gender of the person causing the crash, the vehicle type, the road type, the road movement during the crash, and the geographical location of the crash.

Key Words: Average run length, Decision trees, Early Outbreak Detection, False Alarms, Recursive Partitioning, Statistical process control

Authors:
Ross Sparks, Ross.Sparks@csiro.au
Chris Okugami, chris.okugami@csiro.au

Editor: Xie, M.  , XIE_MIN@NUS.EDU.SG

NOTE: This paper was revised on March 30, 2010.

READING THE ARTICLE: You can read the article in portable document (.pdf) format (440023 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 1932 times since January 9, 2009.


Return to the InterStat Home Page.