Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Discussion of ‘Semiparametric Bayesian optimal replacement policies: application to railroad tracks’ by Merrick and Soyer

Discussion of ‘Semiparametric Bayesian optimal replacement policies: application to railroad... Merrick and Soyer propose a Bayesian semiparametric approach to select optimal replacement policies for maintained mechanical systems. The paper is nicely written and takes the reader step by step through the construction of the priors and derivation of the respective marginal posteriors and the utility function. The data augmentation step is also explained in quite some detail, first in particular cases and then in general, which makes the whole construction easier to grasp. They illustrate their methodology throughout by an application to railroad track maintenance scheduling, but the results are obviously transferable to other systems with similar features. However, we do not see the direct relevance of their approach to software reliability as indicated in the conclusions. The reason is that the applications in software reliability are mostly not connected to maintenance and optimal scheduling of repair actions, but to release testing with immediate repairs.Merrick and Soyer address two issues not usually treated in the statistical literature, namely, how to define a statistical model that does not impose too many constraints on the characteristics of failures in the system (such that the underlying model for failures does not have to be fully specified a priori) and how to include http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Stochastic Models in Business and Industry Wiley

Discussion of ‘Semiparametric Bayesian optimal replacement policies: application to railroad tracks’ by Merrick and Soyer

Download PDF
2 pages

Loading next page...
 
/lp/wiley/discussion-of-semiparametric-bayesian-optimal-replacement-policies-LsiMhooSsQ

References (7)

Publisher
Wiley
Copyright
Copyright © 2017 John Wiley & Sons, Ltd.
ISSN
1524-1904
eISSN
1526-4025
DOI
10.1002/asmb.2265
Publisher site
See Article on Publisher Site

Abstract

Merrick and Soyer propose a Bayesian semiparametric approach to select optimal replacement policies for maintained mechanical systems. The paper is nicely written and takes the reader step by step through the construction of the priors and derivation of the respective marginal posteriors and the utility function. The data augmentation step is also explained in quite some detail, first in particular cases and then in general, which makes the whole construction easier to grasp. They illustrate their methodology throughout by an application to railroad track maintenance scheduling, but the results are obviously transferable to other systems with similar features. However, we do not see the direct relevance of their approach to software reliability as indicated in the conclusions. The reason is that the applications in software reliability are mostly not connected to maintenance and optimal scheduling of repair actions, but to release testing with immediate repairs.Merrick and Soyer address two issues not usually treated in the statistical literature, namely, how to define a statistical model that does not impose too many constraints on the characteristics of failures in the system (such that the underlying model for failures does not have to be fully specified a priori) and how to include

Journal

Applied Stochastic Models in Business and Industry – Wiley

Published: Sep 1, 2017

There are no references for this article.