Use of the Kolmogorov–Smirnov test for gamma process

This article deals with the use of the Kolmogorov–Smirnov test for comparing an observed gamma process with a reference process or for comparing two observed gamma processes. In the case of observed processes with periodic inspections, the Kolmogorov–Smirnov test can be applied directly. It is pointed out that, from the power of test point of view, the length of monitoring is more important than the number of observations. In the case of observed processes with aperiodic inspections, a method for building an empirical cumulative distribution required for the test is proposed. It consists of generating equal time increment observations from the original sample thanks to the gamma bridge. A sort of the original observations for improving the power of the test is proposed. Simulations have proved the feasibility of the approach and shown that when the variance of the time increments increases, the power of the test decreases only a little. An illustrative example showing the interest of such a test for maintenance is described.

[1]  Jan M. van Noortwijk,et al.  A survey of the application of gamma processes in maintenance , 2009, Reliab. Eng. Syst. Saf..

[2]  On an asymptotic goodness-of-fit test for a two-parameter gamma-distribution , 1996 .

[3]  M. Abdel-Hameed Inspection and maintenance policies of devices subject to deterioration , 1987, Advances in Applied Probability.

[4]  E Grall-Maës On a goodness of fit test for gamma process: Comparison of an observed process with a reference , 2011 .

[5]  W. Feller On the Kolmogorov–Smirnov Limit Theorems for Empirical Distributions , 1948 .

[6]  F. Massey The Kolmogorov-Smirnov Test for Goodness of Fit , 1951 .

[7]  Kyung S. Park,et al.  Optimal replacement of an item subject to cumulative damage under periodic inspections , 1997 .

[8]  G. S. Mudholkar,et al.  A gamma goodness-of-fit test based on characteristic independence of the mean and coefficient of variation , 2008 .

[9]  D. Darling,et al.  The Cramer-Smirnov Test in the Parametric Case , 1955 .

[10]  T. W. Anderson,et al.  Asymptotic Theory of Certain "Goodness of Fit" Criteria Based on Stochastic Processes , 1952 .

[11]  N. Smirnov Table for Estimating the Goodness of Fit of Empirical Distributions , 1948 .

[12]  Ram C. Tripathi,et al.  Statistical tests involving several independent gamma distributions , 1993 .

[13]  Hongzhou Wang,et al.  A survey of maintenance policies of deteriorating systems , 2002, Eur. J. Oper. Res..

[14]  G. Jogesh Babu,et al.  Goodness-of-fit tests when parameters are estimated , 2004 .

[15]  Kyung S. Park,et al.  Optimal continuous-wear limit replacement under periodic inspections , 1988 .

[16]  T. W. Anderson On the Distribution of the Two-Sample Cramer-von Mises Criterion , 1962 .

[17]  Rommert Dekker,et al.  On the impact of optimisation models in maintenance decision making: the state of the art , 1998 .

[18]  J. Gurland,et al.  Goodness of Fit Tests for the Gamma and Exponential Distributions , 1972 .

[19]  Rommert Dekker,et al.  Applications of maintenance optimization models : a review and analysis , 1996 .

[20]  Athanassios N. Avramidis,et al.  Efficient simulation of gamma and variance-gamma processes , 2003, Proceedings of the 2003 Winter Simulation Conference, 2003..

[21]  H. Lilliefors On the Kolmogorov-Smirnov Test for Normality with Mean and Variance Unknown , 1967 .

[22]  Rose Baker,et al.  Some new tests of the power law process , 1996 .

[23]  Min Xie,et al.  A simple goodness-of-fit test for the power-law process, based on the Duane plot , 2003, IEEE Trans. Reliab..