Abstract
In this paper age replacement (AR) and opportunity-based age replacement (OAR) for a unit are considered, based on a one-cycle criterion, both for a known and unknown lifetime distribution. In the literature, AR and OAR strategies are mostly based on a renewal criterion, but in particular when the lifetime distribution is not known and data of the process are used to update the lifetime distribution, the renewal criterion is less appropriate and the one-cycle criterion becomes an attractive alternative. Conditions are determined for the existence of an optimal replacement age T * in an AR model and optimal threshold age T opp * in an OAR model, using a one-cycle criterion and a known lifetime distribution. In the optimal threshold age T opp *, the corresponding minimal expected costs per unit time are equal to the expected costs per unit time in an AR model. It is also shown that for a lifetime distribution with increasing hazard rate, the optimal threshold age is smaller than the optimal replacement age. For unknown lifetime distribution, AR and OAR strategies are considered within a nonparametric predictive inferential (NPI) framework. The relationship between the NPI-based expected costs per unit time in an OAR model and those in an AR model is investigated. A small simulation study is presented to illustrate this NPI approach.
Similar content being viewed by others
References
Augustin T and Coolen FPA (2004). Nonparametric predictive inference and interval probability . J Stat Plann Inference 124: 251–272.
Aven T and Jensen U (1999). Stochastic Models in Reliability . Wiley: New York.
Barlow RE and Proschan F (1965). Mathematical Theory of Reliability . Wiley: New York.
Coolen FPA (2006). On nonparametric predictive inference and objective Bayesianism . J Logic Language Inform 15: 21–47.
Coolen FPA and Coolen-Schrijner P (2003). A nonparametric predictive method for queues . Eur J Opl Res 145: 425–442.
Coolen FPA and Yan KJ (2004). Nonparametric predictive inference with right-censored data . J Stat Plann Inference 126: 25–54.
Coolen FPA, Coolen-Schrijner P and Yan KJ (2002). Nonparametric predictive inference in reliability . Reliab Eng Syst Safety 78: 185–193.
Coolen-Schrijner P and Coolen FPA (2004a). Nonparametric predictive inference for age replacement with a renewal argument . Qual Reliab Eng Int 20: 203–215.
Coolen-Schrijner P and Coolen FPA (2004b). Adaptive age replacement strategies based on nonparametric predictive inference . J Opl Res Soc 55: 1281–1297.
Coolen-Schrijner P and Coolen FPA (2006). On optimality criteria for age replacement . Proc Inst Mech Eng Part O: J Risk Reliab 220: 21–28.
Coolen-Schrijner P and Coolen FPA (2007). Nonparametric adaptive age replacement with a one-cycle criterion . Reliab Eng Syst Safety 92: 74–84.
Coolen-Schrijner P, Coolen FPA and Shaw SC (2006). Nonparametric adaptive opportunity-based age replacement strategies . J Opl Res Soc 57: 63–81.
Dekker R and Dijkstra MC (1992). Opportunity-based age replacement: Exponentially distributed times between opportunities . Naval Res Logist 39: 175–190.
Gautschi W and Cahill WF (1964). Exponential integral and related functions . In: Abramowitz M. and Stegun I.A. (eds). Handbook of Mathematical Functions with Formula, Graphs, and Mathematical Tables. National Bureau of Standards: Washington, DC, pp 227–251.
Hill BM (1968). Posterior distribution of percentiles: Bayes' theorem for sampling from a population . J Am Stat Assoc 63: 677–691.
Jhang JP and Sheu SH (1999). Opportunity-based age replacement policy with minimal repair . Reliab Eng Syst Safety 64: 339–344.
Mazzuchi TA and Soyer R (1996). A Bayesian perspective on some replacement strategies . Reliab Eng Syst Safety 51: 295–303.
Sheu SH, Yeh RH, Lin YB and Juan MG (1999). A Bayesian perspective on age replacement with minimal repair . Reliab Eng Syst Safety 65: 55–64.
Sheu SH, Yeh RH, Lin YB and Juan MG (2001). A Bayesian approach to an adaptive preventive maintenance model . Reliab Eng Syst Safety 71: 33–44.
Acknowledgements
This research was started when the second author was a Post-Doctoral Research Assistant at Durham University, supported by the UK Engineering and Physical Research Sciences Council, grant GR/R92530/01. The simulations and computations were performed using the statistical package R, available from www.r-project.org.
Author information
Authors and Affiliations
Corresponding author
Additional information
Dedicated to the memory of Pauline Coolen-Schrijner, who died on 23 April 2008.
Appendix
Appendix
Proof of Theorem 1
It follows from that E X|Y [C opp(X,Y,T)]=E X [C(X,T+Y)] where E X|Y [·] denotes the expectation calculated with respect to the conditional distribution of X given Y. Hence, E X,Y [C opp(X,Y,T)] can be written as
where f Y (y) denotes the pdf of Y. Integrating (A.1) by parts gives
where S Y (y) denotes the survival function of Y. The result follows as S Y (u−T)=eλTe−λu. □
Proof of Theorem 2
Differentiating (6) with respect to T gives
and, consequently, for T opp * to be a stationary point of E X,Y [C opp(X,Y,T)] it must satisfy (7). From (6) it follows that T opp * is a stationary point if E X,Y [C opp(X,Y,T opp *)]=E X [C(X,T opp *)]. Differentiating (A.2) with respect to T gives
Using (7), it follows that for T opp * to be a minimum, the condition E X ′[C(X,T opp *)]<0 must be satisfied. Using (2), E X ′[C(X,T opp *)]<0 is equivalent to (8). □
Proof of Theorem 3
If h X (T) is monotonically increasing, then the AR model has a unique minimum T * which can be obtained by solving T * h X (T *)=c p/(c f−c p). Then E X ′[C(X,T *)]=0 and E X ′[C(X,T)]>0 for T>T *. Hence, for T⩾T *,
so that from (6), E X,Y [C opp(X,Y,T)]>E X [C(X,T)]. In an AR model when T↓0, that is, a unit will be replaced as soon as it is installed, E X [C(X,T)] becomes infinite (see also (1)). However, in an OAR model when T↓0, that is, a unit will be replaced upon the first opportunity after T = 0,E X,Y [C opp(X,Y,0)] is finite (see also (5)). Hence, there is at least one solution to E X,Y [C opp(X,Y,T)]=E X [C(X,T)] and this must occur in the interval (0,T *). If h X (T) is monotonically increasing, then for T in this interval, Th X (T)<c p/(c f−c p) holds, so that from Theorem 2, any solution to E X,Y [C opp(X,Y,T)]=E X [C(X,T)] must correspond to a local minimum and is thus unique. Consequently, this unique minimum T opp * satisfies both (10) and (11). □
Proof of Theorem 4
From Theorem 1 and (2) it follows that
For T∈(0,x 1), using (14) and (22), it follows that
From (14), (15) and (22), it follows that
Using (22), it follows that
and similarly,
Noting that
it follows that
Substituting (A.6), (A.7) and (A.9) into (A.5) and the result together with (A.4) into (A.3) gives (24).
For T∈[x j ,x j+1),j=1,…,n, it follows from (14) and (22) that
From (14) and (15), it follows that
so that with (A.8), it can be deduced that
Substituting (A.12) into (A.11), and substituting the result together with (A.10) into (A.3) yields (25). □
Rights and permissions
About this article
Cite this article
Coolen-Schrijner, P., Shaw, S. & Coolen, F. Opportunity-based age replacement with a one-cycle criterion. J Oper Res Soc 60, 1428–1438 (2009). https://doi.org/10.1057/jors.2008.99
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1057/jors.2008.99