Abstract
This article presents a statistical methodology that may be used in quantifying and optimizing the efficacy of predictive systems used in proactive maintenance. The methodology was developed with regard to a particular tool employed by Bellcore's clients, but is believed to have application to the use of other similar predictive systems. As there are a large number of factors driving trouble reports and typical data sets tend to be highly noisy, until recently there were no clear statistics about how much predictive systems reduce trouble reports. Moreover, there were no clear guidelines about how these systems should be used in an optimum manner. The quantification of the performance of such systems requires careful modeling of the process governing the trouble reports. In this article, by using Generalized Linear Models (GLM), we studied the effect of these tools on Code 4 trouble reports. We also discuss the use of experimental designs in this context to determine the optimum use of the tool under different environmental conditions.
Similar content being viewed by others
References
P. McCullagh and J.A. Neider,Generalized Linear Models (Chapman and Hall, 1983).
T.J. Santner and D.E. Duffy,The Statistical Analysis of Discrete Data (Springer, 1989).
D.M. Gay and R.E. Welsch, Maximum likelihood and quasi-likelihood for nonlinear exponential family regression models, J. Amer. Stat. Assoc. 83(1988)990–998.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Weerahandi, S., Kurien, T.V. & Sadrian, A. Estimating and optimizing the efficacy of predictive systems in proactive maintenance. Telecommunication Systems 6, 315–327 (1996). https://doi.org/10.1007/BF02114301
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02114301