디리슈레 혼합모형을 이용한 함정 전투체계 부품의 고장시간 분포 추정
An Application of Dirichlet Mixture Model for Failure Time Density Estimation to Components of Naval Combat System
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Reliability analysis of the components frequently starts with the data that manufacturer provides. If enough failure data are collected from the field operations, the reliability should be recomputed and updated on the basis of the field failure data. However, when the failure time record for a component contains only a few observations, all statistical methodologies are limited. In this case, where the failure records for multiple number of identical components are available, a valid alternative is combining all the data from each component into one data set with enough sample size and utilizing the useful information in the censored data. The ROK Navy has been operating multiple Patrol Killer Guided missiles (PKGs) for several years. The Korea Multi-Function Control Console (KMFCC) is one of key components in PKG combat system. The maintenance record for the KMFCC contains less than ten failure observations and a censored datum. This paper proposes a Bayesian approach with a Dirichlet mixture model to estimate failure time density for KMFCC. Trends test for each component record indicated that null hypothesis, that failure occurrence is renewal process, is not rejected. Since the KMFCCs have been functioning under different operating environment, the failure time distribution may be a composition of a number of unknown distributions, i.e. a mixture distribution, rather than a single distribution. The Dirichlet mixture model was coded as probabilistic programming in Python using PyMC3. Then Markov Chain Monte Carlo (MCMC) sampling technique employed in PyMC3 probabilistically estimated the parameters’ posterior distribution through the Dirichlet mixture model. The simulation results revealed that the mixture models provide superior fits to the combined data set over single models.
Abstract
Reliability analysis of the components frequently starts with the data that manufacturer provides. If enough failure data are collected from the field operations, the reliability should be recomputed and updated on the basis of the field failure data. However, when the failure time record for a component contains only a few observations, all statistical methodologies are limited. In this case, where the failure records for multiple number of identical components are available, a valid alternative is combining all the data from each component into one data set with enough sample size and utilizing the useful information in the censored data. The ROK Navy has been operating multiple Patrol Killer Guided missiles (PKGs) for several years. The Korea Multi-Function Control Console (KMFCC) is one of key components in PKG combat system. The maintenance record for the KMFCC contains less than ten failure observations and a censored datum. This paper proposes a Bayesian approach with a Dirichlet mixture model to estimate failure time density for KMFCC. Trends test for each component record indicated that null hypothesis, that failure occurrence is renewal process, is not rejected. Since the KMFCCs have been functioning under different operating environment, the failure time distribution may be a composition of a number of unknown distributions, i.e. a mixture distribution, rather than a single distribution. The Dirichlet mixture model was coded as probabilistic programming in Python using PyMC3. Then Markov Chain Monte Carlo (MCMC) sampling technique employed in PyMC3 probabilistically estimated the parameters’ posterior distribution through the Dirichlet mixture model. The simulation results revealed that the mixture models provide superior fits to the combined data set over single models.
참고문헌 (22)
- Austin, P. C., "The Number of Subjects per Variable Required in Linear Regression Analyses", Journal of Clinical Epidemiology, 68, pp.627~636 (2015)
- Bendell, T., "An Overview of Collection, Analysis, and Reliability Data in the Process Industries", IEEE Transactions on Reliability, 37, pp.132~137 (1998)
- 조형준, "와이블분포 하에서 베이지안 기법과 전통적 기법 간의 신뢰도 추정 정확도 비교", 대한산업공학회지, 42(4), pp.256~262 (2016), 대한산업공학회[DOI]
- Introduction to the Dirichlet Distribution and Related Process, "Frigyik, B. A.", University of Washington, ~, pp.UWEETR-2010-0006 (2010)
- Garmabaki, A. H. S., "A Reliability Decision Framework for Multiple Repairable Units", Reliability Engineering and System Safety, 150, pp.78~88 (2016)
- Garmabaki, A. H. S., "Reliability Modelling of Multiple Repairable Units", Quality and Reliability Engineering International, 32, pp.2329~2343 (2016)
- 조인탁, "항공전자장비의 운용자료 분석을 통한 신뢰성 성장 연구", 한국산업경영시스템학회지, 36(4), pp.100~108 (2013), 한국산업경영시스템학회
- 김명수, "수리가능 체계의 필드 데이터 분석", 대한설비관리학회지, 22(3), pp.3~10 (2017), 대한설비관리학회
- 김영진, "야전 치명고장자료를 이용한 함정전투체계 신뢰성 분석 및 활용 방안", 한국산업경영시스템학회지, 39(4), pp.49~59 (2016), 한국산업경영시스템학회
- Kvaloy, J.T., "A Class of Tests for Trend in Time Censored Recurrent Event Data", Technometrics (2019)[DOI]
- Lawless, J., "Testing for Monotone Trend in Recurrent Event Processes", Journal of Technometrics, 54, pp.147~158 (2012)
- 이성환, "수리 가능 시스템의 신뢰성 분석 절차 및 사례 연구", 한국군사과학기술학회지, 9(2), pp.51~59 (2006), 한국군사과학기술학회
- Lewis, P.A.W., "Reliability and Biometry", pp.163~182 (1974), SIAM
- Louit, D. M., "A Practical Procedure for the Selection of Time-to-Failure Models based on the Assessment of Trends in Maintenance Data", Reliability Engineering and System Safety, 94, pp.1618~1628 (2009)
- Miller, J.W., "A Simple Example of Dirichlet Process Mixture Inconsistency for the Number of Components", Advances in Neural Information Processing Systems, pp.199~206 (2013)
- Miller, J. W., "Inconsistency of Pitman-Yor Process Mixtures for the Number of Components", Journal of Machine Learning Research, 15(1), pp.3333~3370 (2014)
- Na, I. Y., "Multi-System Reliability Trend Analysis Model using Incomplete Data with Application to Tank Maintenance", Quality and Reliability Engineering International, 33, pp.2385~2395 (2017)
- "NIST/SEMATECH, e-Handbook of Statistical Methods" (2012)[URL]
- Peduzzi, P., "A Simulation Study of the Number of Events per Variable in Logistic Regression Analysis", Journal of Clinical Epidemiology, 49, pp.1373~1379 (1996)
- Percy, D. F., "Setting Preventive Maintenance Schedules When Data are Sparse", International Journal of Production Economics, 51, pp.223~234 (1997)
- Salvatier, J., "Probabilistic Programming in Python using PyMC3", PeerJ Computer Science, 2 (2016)
- 서양우, "필드데이터 기반의 유도탄 신뢰도 예측", 신뢰성 응용연구, 18(3), pp.250~259 (2018), 한국신뢰성학회[DOI]