Abstract
This article presents a review of two popular methods for temporal surveillance and proposes a general framework for spatial and spatiotemporal surveillance based on likelihood ratio statistics. It is shown that the cumulative sum and Shiryayev-Roberts statistics are special cases under such a general framework. The efficiencies of some surveillance methods are compared for the detection of clusters of high incidence rates in spatial and spatiotemporal applications using both Monte Carlo simulations and a real example.
Original language | English |
---|---|
Pages (from-to) | 724-743 |
Number of pages | 20 |
Journal | IIE Transactions (Institute of Industrial Engineers) |
Volume | 44 |
Issue number | 9 |
DOIs | |
Publication status | Published - 2012 Sept 1 |
Externally published | Yes |
Bibliographical note
Funding Information:We are grateful to the Associate Editor and two anonymous referees for their insightful suggestions that significantly improved the quality of this article. Kwok-Leung Tsui’s research was supported in part by NSF project CMMI-0927592 and Hong Kong RGC General Research Fund (121410) and Food and Health Bureau RFCID Fund (11101262). Wei Jiang’s work was supported in part by HKUSTRPC10EG15 and NSFC 71172131. Bill Woodall’s work was partially supported by NSF grant CMMI-0927323. Sections 4, 5, and 6 of this article contain a portion of the contents in the Ph.D. dissertation by Sung Won Han, which followed the policy and guidelines for Ph.D. dissertations at Georgia Institute of Technology. All copyright of the content is held by IIE Transactions.
Keywords
- Change-point detection
- Shiryayev-Roberts methods
- cumulative sum chart
- detection delay
- scan statistics
- spatial clusters
ASJC Scopus subject areas
- Industrial and Manufacturing Engineering