Generalised proportional intensities models for repairable systems

D. Percy, B. Alkali

Research output: Contribution to journalArticle

Abstract

Based upon the non-homogeneous Poisson process as recommended by Ascher & Feingold (1984), we investigate suitable models for describing the inter-failure times of complex repairable systems. Specifically, we modify and develop the proportional intensities model (PIM) introduced by Cox (1972b) for this purpose. We illustrate the suitability of these models on hypothetical data taken from the first of these two books. Having identified potential benefits from this approach, we extend the PIM to introduce a new class of generalized proportional intensities models (GPIM), which allow for the inclusion of preventive maintenance (PM) and predictor variables. We discuss the properties and variations of GPIM and comment on similarities and differences between these and other proposed models for complex repairable systems. We also demonstrate the application of simple GPIM to a published data set that was collected from the petroleum industry, using the programming language Fortran and the mathematical software Mathcad. Finally, we consider how the analysis can be improved and extended for scheduling PM actions in practice.
Original languageEnglish
Pages (from-to)171-185
Number of pages14
JournalIMA Journal of Management Mathematics
Volume17
Issue number2
Publication statusPublished - 2006

Fingerprint

Repairable System
Directly proportional
Preventive Maintenance
Preventive maintenance
Model
Large scale systems
Complex Systems
Mathematical Software
Non-homogeneous Poisson Process
Repairable system
Petroleum industry
Petroleum
Failure Time
Computer programming languages
Programming Languages
Predictors
Inclusion
Scheduling
Industry

Keywords

  • complex repairable system
  • corrective maintenance
  • preventive maintenance
  • generalized proportional intensities models

Cite this

@article{5337946330ea47fea25e2367d37d59b5,
title = "Generalised proportional intensities models for repairable systems",
abstract = "Based upon the non-homogeneous Poisson process as recommended by Ascher & Feingold (1984), we investigate suitable models for describing the inter-failure times of complex repairable systems. Specifically, we modify and develop the proportional intensities model (PIM) introduced by Cox (1972b) for this purpose. We illustrate the suitability of these models on hypothetical data taken from the first of these two books. Having identified potential benefits from this approach, we extend the PIM to introduce a new class of generalized proportional intensities models (GPIM), which allow for the inclusion of preventive maintenance (PM) and predictor variables. We discuss the properties and variations of GPIM and comment on similarities and differences between these and other proposed models for complex repairable systems. We also demonstrate the application of simple GPIM to a published data set that was collected from the petroleum industry, using the programming language Fortran and the mathematical software Mathcad. Finally, we consider how the analysis can be improved and extended for scheduling PM actions in practice.",
keywords = "complex repairable system, corrective maintenance, preventive maintenance, generalized proportional intensities models",
author = "D. Percy and B. Alkali",
year = "2006",
language = "English",
volume = "17",
pages = "171--185",
journal = "IMA Journal of Management Mathematics",
issn = "1471-678X",
number = "2",

}

Generalised proportional intensities models for repairable systems. / Percy, D.; Alkali, B.

In: IMA Journal of Management Mathematics, Vol. 17, No. 2, 2006, p. 171-185.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Generalised proportional intensities models for repairable systems

AU - Percy, D.

AU - Alkali, B.

PY - 2006

Y1 - 2006

N2 - Based upon the non-homogeneous Poisson process as recommended by Ascher & Feingold (1984), we investigate suitable models for describing the inter-failure times of complex repairable systems. Specifically, we modify and develop the proportional intensities model (PIM) introduced by Cox (1972b) for this purpose. We illustrate the suitability of these models on hypothetical data taken from the first of these two books. Having identified potential benefits from this approach, we extend the PIM to introduce a new class of generalized proportional intensities models (GPIM), which allow for the inclusion of preventive maintenance (PM) and predictor variables. We discuss the properties and variations of GPIM and comment on similarities and differences between these and other proposed models for complex repairable systems. We also demonstrate the application of simple GPIM to a published data set that was collected from the petroleum industry, using the programming language Fortran and the mathematical software Mathcad. Finally, we consider how the analysis can be improved and extended for scheduling PM actions in practice.

AB - Based upon the non-homogeneous Poisson process as recommended by Ascher & Feingold (1984), we investigate suitable models for describing the inter-failure times of complex repairable systems. Specifically, we modify and develop the proportional intensities model (PIM) introduced by Cox (1972b) for this purpose. We illustrate the suitability of these models on hypothetical data taken from the first of these two books. Having identified potential benefits from this approach, we extend the PIM to introduce a new class of generalized proportional intensities models (GPIM), which allow for the inclusion of preventive maintenance (PM) and predictor variables. We discuss the properties and variations of GPIM and comment on similarities and differences between these and other proposed models for complex repairable systems. We also demonstrate the application of simple GPIM to a published data set that was collected from the petroleum industry, using the programming language Fortran and the mathematical software Mathcad. Finally, we consider how the analysis can be improved and extended for scheduling PM actions in practice.

KW - complex repairable system

KW - corrective maintenance

KW - preventive maintenance

KW - generalized proportional intensities models

UR - http://dx.doi.org/10.1093/imaman/dpi034

M3 - Article

VL - 17

SP - 171

EP - 185

JO - IMA Journal of Management Mathematics

JF - IMA Journal of Management Mathematics

SN - 1471-678X

IS - 2

ER -