Availability analysis for the quasi-renewal process

  • Authors:
  • Ian J. Rehmert;Joel A. Nachlas

  • Affiliations:
  • Servicing Operations Analytics, The Hong Kong and Shanghai Banking Corporation North America's Consumer and Mortgage Lending, Elmherst, IL;Grado Department of Industrial and Systems Engineering, Virginia Tech, Blacksburg, VA

  • Venue:
  • IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans - Special section: Best papers from the 2007 biometrics: Theory, applications, and systems (BTAS 07) conference
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Historically, the behaviors of repairable systems were usually modeled under the assumption that repair implied system renewal. Availability functions were then constructed using renewal functions. Often, equipment is not renewed by repair, and for equipment that is not renewed, existing models fail to capture the key features of their behavior--ongoing degradation. More recently, nonrenewal models have been proposed to reflect the fact that equipment is usually not as good as new following maintenance. A wide variety of such models have been defined. They are usually called imperfect repair models. These models have the advantage that they are more realistic but they are also more complicated; therefore, analytical results for the models have been limited. In this paper, one of the nonrenewal models is analyzed, and an approach for obtaining a detailed measure of equipment performance, the point availability, is presented. The ultimate point availability function must be approximated numerically. Nevertheless, the analysis does lead to the time-dependent measure for a variety of possible distribution models. This paper contains two contributions to the study of repairable equipment performance. First, the models analyzed include both stochastic equipment deterioration and stochastically degrading repair performance over multiple operating intervals. Second, the analytical approach to obtaining the point availability function and its approximation is based on the combined analysis of operation-time- and downtime-based formulations of the system availability. This analytical approach to availability computation has not been used previously and is quite effective.