식ㆍ의약 위해 감시체계(K-RISS)의 우선순위 평가를 위한 시계열 구조변화 기반 기준선 설정 모델 개발
Development of a Baseline Setting Model Based on Time Series Structural Changes for Priority Assessment in the Korea Risk Information Surveillance System (K-RISS)
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Background: The Korea Risk Information Surveillance System (K-RISS) was developed to enable the early detection of food and drug safety-related issues. Its goal is to deliver real-time risk indicators generated from ongoing food and drug risk monitoring. However, the existing K-RISS system suffers under several limitations. Objectives: This study aims to augment K-RISS with more detailed indicators and establish a severity standard that takes into account structural changes in the daily time series of K-RISS values. Methods: First, a Delphi survey was conducted to derive the required weights. Second, a control chart, commonly used in statistical process controls, was utilized to detect outliers and establish caution, attention, and serious levels for K-RISS values. Furthermore, Bai and Perron’s method was employed to determine structural changes in K-RISS time series. Results: The study incorporated ‘closeness to life’ and ‘sustainability’ indicators into K-RISS. It obtained the necessary weights through a survey of experts for integrating variables, combining indicators by data source, and aggregating sub K-RISS values. We defined caution, attention, and serious levels for both average and maximum values of daily K-RISS. Furthermore, when structural changes were detected, leading to significant variations in daily K-RISS values according to different periods, the study systematically verified these changes and derived respective severity levels for each period. Conclusions: This study enhances the existing K-RISS system and introduces more advanced indicators. K-RISS is now more comprehensively equipped to serve as a risk warning index. The study has paved the way for an objective determination of whether the food safety risk index surpasses predefined thresholds through the application of severity levels.
Abstract
Background: The Korea Risk Information Surveillance System (K-RISS) was developed to enable the early detection of food and drug safety-related issues. Its goal is to deliver real-time risk indicators generated from ongoing food and drug risk monitoring. However, the existing K-RISS system suffers under several limitations. Objectives: This study aims to augment K-RISS with more detailed indicators and establish a severity standard that takes into account structural changes in the daily time series of K-RISS values. Methods: First, a Delphi survey was conducted to derive the required weights. Second, a control chart, commonly used in statistical process controls, was utilized to detect outliers and establish caution, attention, and serious levels for K-RISS values. Furthermore, Bai and Perron’s method was employed to determine structural changes in K-RISS time series. Results: The study incorporated ‘closeness to life’ and ‘sustainability’ indicators into K-RISS. It obtained the necessary weights through a survey of experts for integrating variables, combining indicators by data source, and aggregating sub K-RISS values. We defined caution, attention, and serious levels for both average and maximum values of daily K-RISS. Furthermore, when structural changes were detected, leading to significant variations in daily K-RISS values according to different periods, the study systematically verified these changes and derived respective severity levels for each period. Conclusions: This study enhances the existing K-RISS system and introduces more advanced indicators. K-RISS is now more comprehensively equipped to serve as a risk warning index. The study has paved the way for an objective determination of whether the food safety risk index surpasses predefined thresholds through the application of severity levels.
참고문헌 (20)
- "News Sentiment Index (NSI) released as experimental statistic"[URL]
- Development of financial instability index and analysis of changes in financial instability factors, "Jeong YS", Korea Institute for International Economic Policy, ~, pp.Report No.: 18-22 (2018)
- A study on development of consumer safety indicators, "Sah JY", Korea Consumer Agency, ~, pp.Report No.:18-20 (2018)
- "Announcement of 2020 Children’s Dietary Safety Index survey results"[URL]
- "Environment and Health Information System (ENHIS)"[URL]
- U.S.-Mexico Border Field Office of the Pan American Health Organization, "Environmental health indicators workshop U.S. - Mexico Border; 2002 Aug 28-30; Mc Allen, U.S", U.S.-Mexico Border Field Office of the Pan American Health Organization, ~ (2002)
- "DRMKC-INFORM"[URL]
- National Health Insurance Service, "2020 national health insurance statistical yearbook", pp.602~611 (2021), National Health Insurance Service, Health Insurance Review and Assessment Service
- Shewhart WA, "Economic control of quality of manufactured product" (1931), D. Van Nostrand Company
- Kim JG, "Research results and trends analysis for EWMA control chart of manufacturing processes", Paper presented at: Korean Society of Safety Management Sciences 2013 Spring Conference, pp.581~591 (2013)
- Roberts SW, "Control chart tests based on geometric moving averages", Technometrics, 1(3), pp.239~250 (1959)
- Montgomery DC., "Introduction to statistical quality control" (2013), Wiley
- Cho JH, "A comparative analysis on the efficiency of monitoring between EWMA and Shewhart chart in instrumental process with autocorrelation", J Korean Soc Ind Syst Eng, 35(4), pp.118~125 (2012)
- Kim YB, "Statistical design of Shewhart control chart with runs rules", J Korean Soc Qual Manag, 36(3), pp.34~44 (2008)
- Andrews DWK, "Tests for parameter instability and structural change with unknown change point", Econometrica, 61(4), pp.821~856 (1993)
- Andrews DWK, "Optimal tests when a nuisance parameter is present only under the alternative", Econometrica, 62(6), pp.1383~1414 (1994)
- Bai J, "Estimating and testing linear models with multiple structural changes", Econometrica, 66(1), pp.47~78 (1998)
- Bai J, "Critical values for multiple structural change tests", Econom J, 6(1), pp.72~78 (2003)
- Liu J, "On segmented multivariate regression", Stat Sin, 7(2), pp.497~525 (1997)
- Lee SI, "Comparison of control charts for individual observations", Korean J Appl Stat, 35(2), pp.203~215 (2022)