Abstract
In this paper, we propose an optimal Bayesian control policy with two sampling intervals minimizing the long-run expected average maintenance cost per unit time for a partially observable deteriorating system. Unlike the previous optimal Bayesian approaches which used periodic sampling models with equidistant intervals, a novel sampling methodology is proposed which is characterized by two sampling intervals and two control thresholds. The deterioration process is modeled as a 3-state continuous time hidden-Markov process with two unobservable operating-states and an observable failure state. At each sampling epoch, the multivariate observation data provides only partial information about the actual state of the system. We start observing the system with a longer sampling interval. If the posterior probability that the system is in the warning state exceeds a warning limit, observations are taken more frequently, i.e., the sampling interval changes to a shorter one, and if the posterior probability exceeds a maintenance limit, the full inspection is performed, followed possibly by preventive maintenance. We formulate the maintenance control problem in a partially observable Markov decision process (POMDP) framework to find the two optimal control limits and two sampling intervals. Also, the mean residual life (MRL) of the system is calculated as a function of the posterior probability. A numerical example is provided and comparison of the proposed scheme with several alternative sampling and maintenance control strategies is carried out.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Abbreviations
- X t :
-
State of the system at time t
- q i j :
-
Instantaneous transition rate
- P i j (t):
-
Transition probability function from state i to state j
- ν i :
-
Exponential distribution parameter of the sojourn time in state i
- ξ :
-
Failure time of the system
- Δ1 :
-
Longer sampling interval
- Δ2 :
-
Shorter sampling interval
- \(Y_{n_1{\Delta }_1}\) :
-
Residual of the observation at time n 1Δ1
- μ i :
-
Mean vector of normally distributed observation in state i
- Σ i :
-
Covariance matrix of normally dis tributed observation in state i
- f(y∣i):
-
Conditional density of the observation vector given the state is i
- W :
-
Warning limit
- M :
-
Maintenance control limit
- (h,I):
-
Inspection state
- PM :
-
Preventive maintenance state
- T I :
-
Inspection time
- T M :
-
Preventive maintenance time
- T F :
-
Failure replacement time
- τ :
-
Expected sojourn time
- C :
-
Expected cost
- K :
-
Number of subintervals of [0,1]
- \({\Pi }_{n_1{\Delta }_1}\) :
-
Posterior probability at time n 1Δ1
- f ξ (t):
-
Probability density of the time to failure
- E(ξ):
-
Expected time to failure
- MRL :
-
Mean residual life of the system
- \(R(t \mid {\Pi }_{n_1{\Delta }_1+n_2{\Delta }_2})\) :
-
Conditional reliability function
- h(t):
-
Hazard rate function
References
Barlow RE, Hunter LC (1960) Optimum preventive maintenance policies. Oper Res 8:90–100
Barlow RE, Hunter LC, Proschan F (1963) Optimum checking procedure. J Soc Ind Appl Math 11:1078–1095
Barlow RE (1975) Statistical theory of reliability and life testing, Probability models. Operation research
Celano G, Costa A, Fichera S (2006), vol 28, pp 966–977
Calabrese JM (1995) Bayesian process control for attributes. Manag Sci 41:637–645
Chen D, Trivedi K (2005) Optimization for condition-based maintenance with semi-markov decision process. Reliab Eng Syst Saf 90:25–29
Dieulle L, Berenguer C, Grall A, Roussignol V (2003) Sequential condition-based maintenance scheduling for a deteriorating system. Eur J Oper Res 150:451–461
Gupta A, Lawsirirat C (2006) Strategically optimum maintenance of monitoringenabled multi-component systems using continuous-time jump deterioration models. J Q Maint Eng 12:306–329
Imhof JP (1961) Computing the distribution of quadratic forms in normal variable. Biometrika 48(3):419–426
Jardine AK, Lin D, Banjevic D (2006) A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mech Syst Sig Process 20:1483–1510
Jiang X, Yu J, Makis V (2012) Optimal Bayesian estimation and control scheme for gear shaft fault detection. Comput Ind Eng 63:754–762
Kadi D, Beaucaire C, Cleroux R (2006) A periodic maintenance model with used equipment and random minimal repair. Nav Res Logist 37:855–865
Kallen MJ, Noortwijk JM (2005) Optimal maintenance decisions under imperfect inspection. Reliab Eng Syst Saf 90:177–185
Kim MJ, Makis V (2009) Optimal maintenance policy for a multi-state deteriorating system with two types of failures under general repair. Comput Ind Eng 57:298–303
Kim MJ, Makis V, Jiang R (2010) Parameter estimation in a condition-based maintenance model. Stat Probab Lett 80:1633–1639
Kim MJ, Jiang R, Makis V, Lee CG (2011) Optimal Bayesian fault prediction scheme for a partially observable system subject to random failure. Eur J Oper Res 214:331–339
Kim MJ, Makis V (2012) Optimal control of partially observable failing system with costly multivariate observations. Stoch Model 28:584–608
Kim MJ, Makis V (2013) Joint optimization of sampling and control of partially observable failing systems. Oper Res 62:777–790
Makis V, Jiang X (2003) Optimal replacement under partial observations. Math Oper Res 28:382–394
Makis V (2008) Multivariate bayesian control chart. Oper Res 56:487–496
Nenes G, Tagaras G (2007) The economically designed two-sided bayesian x control chart. Eur J Oper Res 183:263–277
Ohnishi M, Kawai H, Mine H (1986) An optimal inspection and replacement policy for a deteriorating system. J Appl Probab 23:973–988
Ou Y, Wu Z, Yu FJ, Shamsuuzaman M (2011) An SPRT control chart with variable sampling interval. Int J Adv Manuf Techno 56:1149–1158
Provost SB, Rudiuk EM (1996) The exact distribution of indefinite quadratic forms in noncentral normal vectors. Annal Inst Stat Math 48:381–394
Reynolds MR (1989) Optimal variable sampling interval control charts. Seq Anal: Des Methods Appl 8:361–379
Reynolds MR, Arnold JC (1989) Optimal one-sided shewhart control charts with variable sampling intervals. Seq Anal 8:51–77
Sheu SH, Griffith W (2002) Extended block replacement policy with shock models and used items. Eur J Oper Res 140:50–60
Tijms HC (1994) Stochastic models—an algorithmic approach, Wiley
Yang SF, Su HC (2007) Adaptive sampling interval cause- selecting control charts. Int J Adv Manuf Technol 31:1169–1180
Yeung T, Cassady C, Schneider K (2008) Simultaneous optimization of control chart and age-based preventive maintenance policies under an economic objective. IIE Trans 40:147–159
Xiang Y, Cassady CR, Pohi E (2012) Optimal maintenance policies for systems subject to a Markovian operating environment. Comput Ind Eng 62:190–197
Xiang Y, Cassady CR, Pohi E (2012) Optimal maintenance policies for systems subject to a Markovian operating environment. Comput Ind Eng 62:190–197
Wang L, Chu J, Mao W (2009) A condition-based replacement and spare provisioning policy for deteriorating systems with uncertain deterioration to failure. Eur J Oper Res 194:184–205
Zhang J, Li Z, Wang Z (2012) A new adaptive control chart for monitoring process mean and variability. Int J Adv Manuf Technol 60:1031–1038
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Naderkhani ZG, F., Makis, V. Optimal condition-based maintenance policy for a partially observable system with two sampling intervals. Int J Adv Manuf Technol 78, 795–805 (2015). https://doi.org/10.1007/s00170-014-6651-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00170-014-6651-4