Models for Perfect Repair
In this chapter, we begin with the study of repairable systems. Specifically, we study probabilistic models for systems which after failure are replaced by a new one exactly. We say then that the system operating state is restored to “as good as new” conditions after failure. In the first approach to this kind of systems, we assume that the system is repaired (replaced) and put into new operation immediately after the failure. The sample information that we analyze consist of a sequence of random variables independent and identically distributed, which represent the time between two consecutive failures. The model that we consider in this situation is a Renewal Process, and the main purpose is to present and compare several ways of nonparametrically estimate the renewal function, that is, the expected number of failures occurring in the system up to a given time t. When the repair times are of interest, the data collected consist of a sequence of alternating up and down periods. The modeling tool indicated in this case is the Alternating Renewal Process, and we will concentrate on estimating the availability function (the probability that the system is functioning at a given time) using nonparametric techniques.
KeywordsRepairable System Kernel Estimator Repair Time Empirical Distribution Function Availability Measure
Section 2.3 of this chapter is an extension into book-length form of the article Nonparametric estimation of the availability in a general repairable system, originally published in Reliability Engineering and System Safety93 (8), 1188–11962 (2008).
The authors express their full acknowledgement of the original publication of the paper in the journal cited above, edited by Elsevier.
- 2.Ascher H, Feingold H (1984) Repairable systems reliability: modelling, inference, misconceptions and their causes. Marcel Dekker Inc, New YorkGoogle Scholar
- 3.Barlow RE, Proschan F (1975) Statistical theory of reliability and life testing. Probability Models. Holt, Rinehart and Winston, New YorkGoogle Scholar
- 6.Cao R (1993) Bootstrapping the mean integrated squared error. J Multivar Anal 45:137–160Google Scholar
- 20.Guillamón A, Navarro J, Ruiz JM (1999) A note on kernel estimators for positive valued random variables. Sankhya: Indian J Stat 61 Ser A (Part 2):276–281Google Scholar
- 21.Hansen BE (2004) Bandwidth selection for nonparametric distribution estimation. Discussion paper, University of Wisconsin, MadisonGoogle Scholar
- 26.Kijima M (1989) Some results for repairable systems with general repair. J Appl Probab 26(1):89–102Google Scholar
- 29.Markovich NM (2004) Nonparametric renewal function estimation and smoothing by empirical data. Preprint ETH, ZuerichGoogle Scholar
- 38.Ruckdeschel P, Kohl M, Stabla T, Camphausen F (2006) S4 classes for distributions. R News 6(2):2–6. http://www.uni-bayreuth.de/departments/math/org/mathe7/DISTR/distr.pdf Google Scholar
- 40.Vanderperre EJ, Makhanov SS (2008) Point availability of a robot with internal safety device. Contemp Eng Sci 1(1):15–25Google Scholar