Presentation 2010-11-19
Inference of Reliability for Degradation Data Based on Stochastic Process Models
Toru KAISE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Methodologies of reliability analysis using stochastic process models for degradation data are proposed in many references. The Bownian motion and the gamma process are used for the analysis. It is possible to use estimation methods for the stochastic models. For example, the methods to estimate parameters are the maximum likelihood, the generalized moment, and the Bayesian methodologies. The information criteria are useful for the model selection. However it is difficult to make application of unified comparisons to the estimation methodologies. In this paper, we handle stochastic process models and estimation methods based on a view point of the Levy process. It is shown that the application of the information criterion EIC make the effects for degradation process analysis, becanse EIC is possible to treat uniformity the selections of the models and the estimations.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Stochastic process / Maximum likelihood / Generalized moment method / Bayes / EIC
Paper # R2010-36
Date of Issue

Conference Information
Committee R
Conference Date 2010/11/12(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Reliability(R)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Inference of Reliability for Degradation Data Based on Stochastic Process Models
Sub Title (in English)
Keyword(1) Stochastic process
Keyword(2) Maximum likelihood
Keyword(3) Generalized moment method
Keyword(4) Bayes
Keyword(5) EIC
1st Author's Name Toru KAISE
1st Author's Affiliation Graduate School of Business, University of Hyogo()
Date 2010-11-19
Paper # R2010-36
Volume (vol) vol.110
Number (no) 298
Page pp.pp.-
#Pages 4
Date of Issue