• DocumentCode
    1020129
  • Title

    Transient analysis of a repairable system, using phase-type distributions and geometric processes

  • Author

    Perez-Ocon, Rafael ; Montoro-Cazorla, Delia

  • Author_Institution
    Fac. de Ciencias, Granada Univ., Spain
  • Volume
    53
  • Issue
    2
  • fYear
    2004
  • fDate
    6/1/2004 12:00:00 AM
  • Firstpage
    185
  • Lastpage
    192
  • Abstract
    The transient behavior of a system with operational and repair times distributed following phase-type distributions is studied. These times are alternated in the evolution of the system, and they form 2 separate geometric processes. The stationary study of this system when the repair times form a renewal process has been made . This paper also considers that operational times are partitioned into two well-distinguished classes successively occupied: good, and preventive. An algorithmic approach is performed to determine the transition probabilities for the Markov process which governs the system, and other performance measures beyond those in are calculated in a well-structured form. The results are applied to a numerical example, and the transient quantities are compared with the ones obtained in the stationary case. The computational implementation of the mathematical expressions formulated are performed using the Matlab program.
  • Keywords
    Markov processes; failure analysis; reliability; stochastic programming; transient analysis; Markov process; Matlab program; algorithmic approach; failure occurrence; geometric process; phase-type distribution; reliability; renewal process; repairable system; transient analysis; transition probabilities; well-structured form; Availability; Exponential distribution; Laplace equations; MATLAB; Markov processes; Partitioning algorithms; Performance evaluation; Probability; Stochastic processes; Transient analysis; Degradation; Markov proccess; phase-type distribution; rate of occurrence of failures; reliability;
  • fLanguage
    English
  • Journal_Title
    Reliability, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0018-9529
  • Type

    jour

  • DOI
    10.1109/TR.2004.829145
  • Filename
    1308662