Abstract
Multivariate CUSUM charts formed over spatial clusters have been used over the last several years to detect emerging disease clusters in spatiotemporal biosurveillance. The control limits for the CUSUM charts are typically calibrated by trial-and-error simulation, but this task can be time-consuming and challenging when the monitoring area is large. This article introduces an analytical method that approximates the control limits and average run length when spatial correlation is not strong. In addition, the practical range of the scan radius in which the approximation method works well is investigated. Also studied is how the outbreak radius and spatial correlation impact the scheme’s outbreak detection performance with respect to two metrics: detection delay and identification accuracy. Experimental results show that the approximation method performs well, making the design of the multivariate CUSUM chart convenient; and higher spatial correlation does not always yield faster detection but often facilitates accurate identification of outbreak clusters.
Additional information
Notes on contributors
Mi Lim Lee
Mi Lim Lee is a Ph.D. Candidate in the H. Milton Stewart School of Industrial and Systems Engineering at the Georgia Institute of Technology. She received her B.S. and M.S. degrees in Industrial Engineering from the Korea Advanced Institute of Science and Technology in 2004 and 2006, respectively. Her research interests include healthcare and public health applications of statistics and simulation. Her e-mail address is [email protected].
David Goldsman
David Goldsman is a Professor in the H. Milton Stewart School of Industrial and Systems Engineering at the Georgia Institute of Technology. His research interests include simulation output analysis, statistical ranking and selection methods, and medical and humanitarian applications of operations research. He was the IIE Board Representative to the Winter Simulation Conference from 2001 to 2009 and he is an IIE Fellow. His e-mail address is [email protected].
Seong-Hee Kim
Seong-Hee Kim is an Associate Professor in the H. Milton Stewart School of Industrial and Systems Engineering at the Georgia Institute of Technology. She served on the Editorial Board of the Simulation Department of IIE Transactions and was an Associate Editor in the simulation area of Operations Research and the OR/simulation area of The American Statistician. She is currently an Associate Editor of the INFORMS Journal on Computing. Her e-mail address is [email protected].
Kwok-Leung Tsui
Kwok-Leung Tsui is Head and Chair Professor in the Department of Systems Engineering and Engineering Management at City University of Hong Kong. Prior to his current position, he was a Professor in the H. Milton Stewart School of Industrial and Systems Engineering at the Georgia Institute of Technology from 1990 to 2011 and he was on the research staff in the Quality Assurance Center at AT&T Bell Labs during 1986–1990. He received his Ph.D. in Statistics from the University of Wisconsin at Madison. He was a recipient of the NSF Young Investigator Award; an (elected) President and Vice President of the American Statistical Association Atlanta Chapter; the Chair of the INFORMS Section in Quality, Statistics, and Reliability; and the Founding Chair of the Section in Data Mining. He is a fellow of American Statistical Association, a U.S. representative in the ISO Technical Committee on Statistical Methods, and a Department Editor of IIE Transactions. His current research interests include health informatics, data mining and surveillance in healthcare and public health, prognostics and health management, calibration and validation of computer models, bioinformatics, process control and monitoring, and robust design and the Taguchi method.