Residual-life estimation for components with non-symmetric priors Academic Article

abstract

• Condition monitoring uses sensory signals to assess the health of engineering systems. A degradation model is a mathematical characterization of the evolution of a condition signal. Our recent research focuses on using degradation models to compute residual-life distributions for degrading components. Residual-life distributions are important for providing probabilistic estimates of failure time for use in maintenance planning and spare parts inventory management. To obtain residual-life distributions, our earlier work assumed the degradation model's stochastic parameters to be normally distributed. This paper investigates the performance of these residual-life distributions when the underlying normality assumptions are not satisfied. The paper also develops methods for estimating residual-life when the stochastic parameters of the degradation model follow more general distributions.

authors

• Lawley, Mark

published proceedings

• IIE Transactions

author list (cited authors)

• Chakraborty, S., Gebraeel, N., Lawley, M., & Wan, H.

citation count

• 35

complete list of authors

• Chakraborty, Santanu||Gebraeel, Nagi||Lawley, Mark||Wan, Hong

publication date

• February 2009

publisher

• Taylor & Francis  Publisher

published in

• IIE Transactions  Journal

keywords

• Bioengineering

Digital Object Identifier (DOI)

• 10.1080/07408170802369409

start page

• 372

end page

• 387

volume

• 41

issue

• 4

URL

• http://dx.doi.org/10.1080/07408170802369409