Estimation of Lifetime Distribution with Covariates Using Online Monitoring  

Oleh:

Yokoyama, Masahiro ; Yamamoto, Watalu ; Suzuki, Kazuyuki

Jenis: Article from Proceeding
Dalam koleksi: 12th ANQ Congress in Singapore, 5-8 Agustus 2014, page 1-10.
Topik: Misspecification; Weibull distribution; Log-normal distribution; Accelerated lifetime model; Covariate
Fulltext: RL2-2.3-P0251.pdf (800.71KB)

Isi artikel

Online monitoring becomes a common tool for keeping products and systems highly reliable in these days. The information to maintain the target product includes usage history, system conditions, and environmental conditions. Nowadays they are monitored and reported in real time and stored as big data. On modeling the failure mechanism statistically, these variables are primary candidates for covariates which affect the failure mechanism. There is literature on modeling the lifetime of products with the accelerated lifetime model on the accumulation of covariate effects which is a nonlinear transformation of the observed lifetime. The existing literature requires the parametric form of lifetime distribution in advance. However such knowledge may be difficult to acquire in advance in some cases. If we assume incorrect distribution, it is called misspecified. This paper proposes a strategy to use log-normal likelihood for the estimation of parameters of covariate effects when the underlying distribution is either Weibull or log-normal. It is derived that the score function of log-normal likelihood is identified as an approximation for Weibull cases. The simulation study shows the relationship among the sample distribution of parameter estimates and underlying distributions. For Weibull cases, if a shape parameter is large, the bias of the resulting estimates is small

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali