Research Article  Open Access
Constantinos C. Karamatsoukis, Epaminondas G. Kyriakidis, "Optimal Maintenance of a Production System with Intermediate Buffers", Mathematical Problems in Engineering, vol. 2012, Article ID 673864, 14 pages, 2012. https://doi.org/10.1155/2012/673864
Optimal Maintenance of a Production System with Intermediate Buffers
Abstract
We consider a productioninventory system that consists of an inputgenerating installation, a production unit and L intermediate buffers. It is assumed that the installation transfers the raw material to buffer and the production unit pulls the raw material from buffer We consider the problem of the optimal preventive maintenance of the installation if the installation deteriorates stochastically with usage and the production unit is always in operative condition. We also consider the problem of the optimal preventive maintenance of the production unit if the production unit deteriorates stochastically with usage and the installation is always in operative condition. Under a suitable cost structure and for given contents of the buffers, it is proved that the averagecost optimal policy for the first (second) problem initiates a preventive maintenance of the installation (production unit) if and only if the degree of deterioration of the installation (production unit) exceeds some critical level. Numerical results are presented for both problems.
1. Introduction
In the present paper, we study two problems, and we generalize the results obtained in two previous papers by Kyriakidis and Dimitrakos [1] and Pavitsos and Kyriakidis [2] that are concerned with the preventive maintenance of a productioninventory system. We consider a manufacturing system (see Figure 1) in which an inputgenerating installation () transfers raw materials to a subsequent production unit (). We assume that buffers have been built between the installation and the production unit. The installation transfers the raw material to the buffer , and the production unit pulls this raw material from the buffer . The buffers have finite capacities.
In the first problem it is assumed that the installation deteriorates stochastically over time, and the production unit is always in operative condition. The deteriorating process for the installation is described by some known transition probabilities between different degrees of deterioration. A discretetime Markov decision model is considered for the optimal preventive maintenance of the installation. The maintenance times are assumed to be geometrically distributed, and the cost structure includes operating costs of the installation, costs for storing the raw materials in the buffers, maintenance costs and costs due to production delay when the installation does not operate or operate partially and the contents of some or all buffers are below some specific levels. It is proved that for fixed contents of the buffers the policy that minimizes the longrun expected average cost per unit time is of controllimit type, that is, it initiates a preventive maintenance of the installation if and only if its degree of deterioration exceeds some critical level. This result generalizes the structural result that was obtained by Kyriakidis and Dimitrakos [1] for the case in which . In the second problem it is assumed that the production unit deteriorates stochastically over time and the installation is always in operative condition. The deteriorating process for the production unit is described by some known transition probabilities between different degrees of deterioration. A discretetime Markov decision model is formulated for the optimal preventive maintenance of the production unit. The maintenance times are assumed to be geometrically distributed, and the cost structure includes operating costs of the production unit, costs for the maintenance of the production unit, storage costs, penalty costs, and costs due to the lost production. It is proved that for fixed contents of the buffers the averagecost optimal policy is again of controllimit type, that is, it initiates a preventive maintenance of the production unit if and only if its degree of deterioration exceeds some critical level. This result generalizes the structural result that was obtained by Pavitsos and Kyriakidis [2] for the case in which .
An example of this system could be a production machine that pulls different parts from buffers and assembles them in order to produce the final product. These parts are transferred by a feeder to the buffers. Note that in the last twenty years a great number of maintenance models for productioninventory systems have been studied (see Van Der Duyn Schouten and Vanneste [3], Meller and Kim [4], Iravani and Duenyas [5], Sloan [6], Yao et al. [7], Rezg et al. [8], Dimitrakos and Kyriakidis [9], Karamatsoukis and Kyriakidis [10], and Hadidi et al. [11]). In these models, the preventive maintenance depends on the working condition of a machine and the level of a subsequent buffer. The first problem that we study in the present paper has its origin in a model introduced by Van Der Duyn Schouten and Vanneste [3]. The states of that model consist of the age of a machine and the content of a subsequent buffer that is fed by the machine. The cost structure included costs due to lost production that were incurred when a repair was performed on the machine and the buffer was empty. The repair times of the machine were assumed to be geometrically distributed. It was proved that, for fixed buffer content, the averagecost optimal policy initiates a preventive maintenance of the machine if and only if its age is greater than or equal to a critical value.
The rest of the paper is organized as follows. In the next section, we describe the problem in which only the installation deteriorates with usage, and we derive the structure of the averagecost optimal policy. In Section 3, we study the case in which only the production unit deteriorates over time, and the structure of the averagecost optimal policy is derived. Numerical results are presented for both problems. In the final section, the main conclusions of the paper are summarized, and we propose topics for future research.
2. The Problem when the Installation Deteriorates Stochastically
We consider a productioninventory system (see Figure 1) which consists of an installation () that supplies the buffer with the raw material and a production unit () which pulls units of the raw material from buffer during one unit of time. It is assumed that the production unit is always in operative condition, and that the installation may fail as time evolves. The buffer , has finite capacity which is equal to units of raw material . As long as the buffer , is not full and the installation is in operative condition, the installation may transfer units of the raw material to buffer during one unit of time and the difference is stored in buffer . As soon as buffer , is filled up the installation reduces its speed from to . The numbers , are assumed to be integers.
We suppose that the installation is inspected at discrete, equidistant time epochs (say every day), and is classified into one of working conditions which describe increasing levels of deterioration. Working condition 0 denotes a new installation (or functioning as good as new), while working condition means that the installation is in failed (inoperative) condition and it cannot transfer the raw materials to the buffers. The intermediate working conditions are operative and are ordered ascendingly to reflect their relative degree of deterioration. The transition probability of moving from working condition at time epoch to working condition at time epoch is equal to . We assume that the probability of eventually reaching the working condition from any initial working condition is nonzero. If at a time epoch the installation is found to be at failure state , a corrective maintenance is mandatory. If at a time epoch the installation is found to be at any working condition , a preventive maintenance may be initiated. When a preventive maintenance is performed, the installation does not operate and it does not transfer any raw material to its buffer. It is assumed that the preventive and corrective repair times (expressed in time units) are geometrically distributed with probability of success and , that is, the probability that they will last time units are equal to and , respectively. When a preventive or a corrective maintenance is performed and the buffer , contains units of raw material , the production unit pulls from buffer during one unit of time units of raw material . Both maintenances bring the installation to its perfect condition 0.
We introduce the state PM to denote the situation that a preventive maintenance is performed on the installation. Then the state space of the system is the set = , where is the state in which is the working condition of the installation and , is the content of buffer . A policy is any rule for choosing actions at each time epoch The possible actions are: action 1 (initiate a preventive maintenance), action 2 (initiate a corrective maintenance), action (transfer raw materials only to those buffers that belong to the nonempty subset of the set . If at a time epoch the installation is found to be at state PM or state , the action 1 or the action 2 is compulsory, respectively. If at a time epoch the installation is found to be at working condition , then we may either choose either action 1 or action . Hence, the number of possible actions in this case is since the number of nonempty subsets of is . If at a time epoch action is chosen and belongs to with , then the content of buffer at next time epoch will be . This increase of the buffer content will happen even if the working condition of the installation at next time epoch is the failure state . A policy is said to be stationary if at each time epoch , it chooses one action which depends only on the current state of the system.
The cost structure of the problem includes operating costs of the installation, storage costs, costs due to the lost production, and maintenance costs. If the working condition of the installation is and the buffer , is not full (or full) the cost of transferring (or units of raw material to buffer during one unit of time is equal to (or ). Therefore, if at a time epoch the working condition of the installation is found to be and action is chosen, then the operating cost until the next time epoch is equal to where corresponds to the buffers that are not full and corresponds to the buffers that are full. We assume that the cost of holding a unit of raw material in buffer for one unit of time is equal to . The cost rates during a preventive and a corrective maintenance of the installation are equal to and , respectively. When a preventive or a corrective maintenance is performed on the installation and all buffers are empty (i.e., , the production unit does not pull any raw material from the buffers. In this case we incur a cost due to production delay that is equal to per unit of time. When , we do not incur any such cost since all buffers contain enough raw materials to satisfy the demands of the production unit for one unit of time. When the inequality is satisfied for some (but not for all) , the demands for raw materials of the production unit for one unit of time are partially satisfied. Therefore, the productivity of the production unit is reduced in the sense that the time for the production of the final products increases, since for one unit of time, some of the raw materials that are needed for the production of the final products are not available. In this case it seems reasonable to assume that the cost rate due to production delay is equal to where is the unavailable quantity of the raw material during one unit of time. Similarly, if at a time epoch the working condition of the installation is found to be and the action is chosen, then the cost due production delay until the time epoch is equal to The following conditions on the cost structure and on the transition probabilities are assumed to be valid.
Condition 1. For the sequences and , , are nondecreasing with respect to . That is, as the working condition of the installation deteriorates, the cost of transferring the raw material to buffer increases.
Condition 2. For , , . That is, the cost of transferring units of raw material to buffer during one unit of time is greater than or equal to the cost of transferring units of raw material to buffer during one unit of time.
Condition 3. . That is, the expected time required for a preventive maintenance is smaller than the expected time required for a corrective maintenance.
Condition 4. . That is, the cost rate of a preventive maintenance does not exceed the cost rate of a corrective maintenance.
Condition 5 (An Increasing Failure Rate Assumption). For each , the function is nondecreasing in , .
A consequence of this condition is that , , where is a random variable representing the next working condition of the installation if its present working condition is . It can be shown (see pages 122123 in Derman [12]) that this condition is equivalent to the following one:
Condition 6. For any nondecreasing function , the quantity , is nondecreasing in .
We consider a discretetime Markov decision process in which we aim to find a stationary policy which minimizes the longrun expected average cost per unit time. Note that for this problem was studied in Kyriakidis and Dimitrakos [1].
2.1. The Structure of the Optimal Policy
Let be a discount factor. The minimum expected nstep discounted cost , where is the initial state, can be found for all , recursively, from the following equations (see chapter 1 in Ross [13]): where in (2.1) is a vector with components in which the jth component equals to , if , while, if , it is equal to and in (2.2), and (2.3) is the vector . The initial condition is , . Note that, if for some values of , (2.1) is valid if is changed to for these values of . Note that the first term in the curly brackets in (2.1) corresponds to the best action among all actions , while the second term corresponds to action 1 (i.e., initiate a preventive maintenance of the installation). The first part of the following lemma is needed to prove that the averagecost optimal policy is of controllimit type for fixed levels of the buffers.
Lemma 2.1. For each , we have that (i), (ii).
Proof. We will prove the lemma by induction on . The lemma is valid for , since . We assume that it is valid for . We will show that it is also valid for . First, we prove part (ii) and then part (i).
Part (ii): Let .
For , we have that
The first inequality follows from Condition 4 and from part (ii) of the induction hypothesis. The second inequality follows from Condition 3 and the inequality which is a consequence of part (i) of the induction hypothesis.
Part (i): We have to show that
Inequality (2.5) is easily verified using (2.1) with and part (ii) above
For and , we have that
The above inequality follows from Condition 1 and the inequality
which is implied by part (i) of the induction hypothesis and Condition 6. Hence (2.6) has been proved for . Similarly, we obtain (2.6) if for some values of .
Since the state space is finite, and the state is accessible from every other state under any stationary policy, it follows that there exist numbers , and a constant that satisfy the averagecost optimality equations (see Corollary 2.5 in [13], page 98). For the states , the optimality equations take the following form: If for some values of , we must change in the above equations to for these values of . In view of part (i) of the above lemma we have the following result.
Corollary 2.2. .
The following proposition gives a characterization of the form of the optimal policy.
Proposition 2.3. For fixed contents () of buffers , there exists a critical working condition such that the policy that minimizes the expected longrun average cost per unit time initiates a preventive maintenance of the installation if and only if its working condition is greater than or equal to .
Proof. Suppose that for some fixed such that , , the optimal policy initiates a preventive maintenance of the installation at state where . This implies that
To show that the optimal policy prescribes a preventive maintenance on the installation at state it is enough to show that
From Conditions 1 and 6 and Corollary 2.2, it follows that the righthand side of (2.12) is greater than or equal to the righthand side of (2.11). Hence (2.11) implies (2.12). The same result is obtained similarly if for some values of .
Remark 2.4. In the above proposition, if for fixed contents of the buffers , then the optimal policy never initiates a preventive maintenance of the installation whenever the buffer , , contains units or raw material .
2.2. Numerical Results
Example 2.5. Suppose that , , , , , , , , , , , , , , , = , = , = , , and , , . It can be readily checked that these probabilities satisfy Condition 5. We computed the optimal policy if is equal to 0.5 or 15.5 by implementing the valueiteration algorithm (see Chapter 3 of Tijms [14]). Our numerical results verify the result of Proposition 2.3. In Table 1 we present the critical numbers ,. In each cell of this table the first number corresponds to and the second number corresponds to . The minimum average cost was found to be 7.49 if which is, as expected, considerably smaller than the minimum average cost if , which was found to be 11.63.

We observe that for and the critical number that corresponds to is smaller than or equal to the critical number that corresponds to . This is intuitively reasonable since if the cost due to production delay takes large values it seems disadvantageous to have all or most of the buffers empty when a maintenance is performed on the installation. Therefore in this case it seems preferable to initiate a preventive maintenance of the installation only if its degree of deterioration is relatively high. We also observe that when and buffer or buffer is empty, the optimal policy in most cases never initiates a preventive maintenance of the installation. For example . It can be also seen from the Table 1 that is not a monotone function with respect to and with respect to for constant and , respectively. Note that, when , the value iteration algorithm gives the optimal action for the operation of the installation. For example, if , , , the optimal action when the working condition of the installation is 3 is to transfer raw material 1 to buffer 1 (i.e., ). If , , , the optimal action when the working condition of the installation is 2 is to transfer raw material 1 to buffer 1 and raw material 2 to buffer 2 (i.e., ).
Example 2.6. Suppose that , , , , , , , , , , , , , , , , , , , , . In Table 2 below we present the minimum average cost obtained by the valueiteration algorithm for and .

From the above table we see that as or increases, the minimum average cost decreases. This is intuitively reasonable because in this example it seems favourable to have buffers with large, capacities since the cost rate due to production delay is relatively large while the probabilities , of successful maintenances and the storage cost rates are relatively small.
3. The Problem when the Production Unit Deteriorates Stochastically
We consider the same productioninventory system (see Figure 1) as the one introduced in the previous section with the following modifications: (i) the installation is always in operative condition while the production unit may experience a failure as time evolves and (ii) as long as the buffer , , is not empty and the production unit is in operative condition, the production unit may pull the raw material from buffer at a constant rate of units of raw material per unit of time. When the buffer is empty and the production unit is in operative condition, the production unit reduces its pull rate from to .
We assume that the production unit is monitored at discrete equidistant time epochs (say every day), and is classified into one of working conditions . We suppose that working condition is better than working condition . Working condition 0 means that the production unit is new (or functioning as good as new), while working condition means that the production unit does not function, and it cannot pull the materials from the buffers. The intermediate working conditions are operative. If the working condition at time epoch is then the working condition at time epoch will be with probability . The probability that the deterioration process of the production unit reaches eventually the failure state from any initial working condition is assumed to be nonzero. If at a time epoch the production unit is found to be at the failure state , a corrective maintenance is compulsory. If it is found to be at any working condition , a preventive maintenance is optional. The production unit does not operate when it is under preventive maintenance, and it does not pull any raw material from its buffer. When a preventive or a corrective maintenance is performed and the buffer , , contains units of raw material j, the installation transfers units of raw material to buffer during one unit of time. The preventive and corrective maintenance times (expressed in time units) are geometrically distributed with probability of success and , respectively. Both maintenances bring the production unit to its perfect condition 0. The state space of the system is the set , where PM represents the situation that the production unit is under a preventive repair. The possible actions are the same as the ones considered in the problem studied in Section 2 with the following modification: action is the action of pulling raw materials only from buffers , . If at a time epoch the production unit is found to be at working condition , then we may choose either the action of initiating a preventive maintenance or action .
The cost structure includes operating costs of the production unit, storage costs, maintenance costs, costs due to production delay, and penalty costs. The storage costs , , and the maintenance costs are defined exactly in the same way as in the problem studied in the previous section. If the working condition of the production unit is and the buffer , , is nonempty (or empty), the cost of pulling (or ) units of raw material from buffer during one unit of time is equal to (or ). We assume that the cost rate due to production delay as long as a maintenance of the production unit lasts is equal to . Therefore, if at a time epoch the action is selected and the content of buffer , , is , then the cost due to production delay until time epoch is equal to where is the unavailable quantity of raw material during one unit of time. A penalty cost per unit time which is equal to , , is also imposed for each unit or raw material that is not stored in buffer during a corrective or a preventive maintenance of the production unit when the buffer is full. This cost is due to the necessary labor for transferring and storing the raw material in another place until the completion of the maintenance. We assume that Conditions 1–5 on the cost structure that we introduced for the problem studied in the previous section are valid if we replace with , with , and with . We consider a discretetime Markov decision process in which we aim to find a stationary policy which minimizes the expected longrun average cost per unit of time. Note that for , the problem was studied in Pavitsos and Kyriakidis [2].
Since the state space is finite and the state is accessible from every other state under any stationary policy, it follows that there exist numbers , and a constant that satisfy the averagecost optimality equations. For the states , , , , the optimality equations have the following form: where If for some values of , we must change in (3.2) to for these values. Note that in (3.2) is a vector with components in which the th component is equal to if , while, if , it is equal to . It is possible to prove that ≤ , , , , using the dynamic programming equation for the corresponding finitehorizon problem. The method is exactly the same as that used for the proof of Corollary 2.2 in the previous section and, therefore, we omit the details. An immediate consequence of the above monotonicity result is that the result of Proposition 2.3 is valid for the problem of the optimal preventive maintenance of the production unit.
3.1. Numerical Examples
Example 3.1. Suppose that , , , , , , , , , , , , , , , , , , , , , , and , , . We computed the optimal policy by implementing the valueiteration algorithm. The minimum average cost was found to be 15.67. In Table 3, we present the critical numbers , ≤ ≤ , ≤ ≤ .
We can see from Table 3 that is not a monotone function with respect to or with respect to , respectively. Note that, when , , , the valueiteration algorithm gives the optimal action for the operation of the production unit. For example, if , the optimal action when the working condition of the installation is is to pull raw material 1 from buffer and raw material 2 from buffer (i.e., ), while, when the working condition is the optimal condition is to pull only raw material 2 from buffer (i.e., ).

Example 3.2. Suppose that , , , , , , , , , , , , , , , , , ,, , and , , . In Figure 2 we present the graph of the minimum average cost as a function of , if , and the graph of the minimum average cost as a function of , if . We observe that , , increases as increases. The increase of the minimum average cost is more intense when increases. This can be explained by the fact that the capacity of buffer is considerably greater than the capacity of buffer .
4. Conclusions and Future Research
We presented two discretetime Markov decision models for the optimal conditionbased preventive maintenance of a production system which consists of two machines and intermediate buffers. The first machine transfers different raw materials to the buffers, and the second machine draws the raw materials from the buffers. The second machine is considered to be a production unit that assembles the raw materials in order to produce the final product. In the first model, it is assumed that only the first machine deteriorates stochastically over time while the production unit is always in operative condition. It is possible to monitor the first machine at discrete equidistant time epochs and to classify it into one working condition that describes its level of deterioration. If the first machine is in failed condition a corrective maintenance must be commenced; otherwise a preventive maintenance may be performed or the action of transferring raw materials to any subset of the set of buffers may be selected. The maintenances bring the first machine to its perfect condition. In the second model it is assumed, that only the production unit deteriorates over time, and the first machine is always in operative machine. It is possible to determine the level of deterioration of the production unit after inspecting it at discrete equidistant time epochs. If the production unit is in failed condition a corrective maintenance must be started; otherwise a preventive maintenance may be initiated or the action of pulling the raw materials from any subset of the set of buffers may be selected. Both maintenances bring the production unit to its perfect condition.
In both models we considered the problem of determining the policy that minimizes the expected longrun average cost per unit time. If the maintenance times are geometrically distributed we proved that, in both models, the optimal policy is of controllimit type, that is, for fixed contents of the buffers it prescribes a preventive maintenance of the first machine or the production unit if and only if its degree of deterioration exceeds some critical level. The proof was achieved through the corresponding finitehorizon problem.
A topic for future research could be a more complicated problem in which the first machine transfers the raw materials to the buffers and the production unit draws them from the buffers in a random manner. Another topic for future research could be the study of the maintenance problems that we would have if the maintenance times are not geometrically distributed but follow some general distributions with suitable conditions.
Acknowledgment
The authors would like to thank an anonymous referee for helpful comments.
References
 E. G. Kyriakidis and T. D. Dimitrakos, “Optimal preventive maintenance of a production system with an intermediate buffer,” European Journal of Operational Research, vol. 168, no. 1, pp. 86–99, 2006. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 A. Pavitsos and E. G. Kyriakidis, “Markov decision models for the optimal maintenance of a production unit with an upstream buffer,” Computers and Operations Research, vol. 36, no. 6, pp. 1993–2006, 2009. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 F. A. van der Duyn Schouten and S. G. Vanneste, “Maintenance optimization of a production system with buffer capacity,” European Journal of Operational Research, vol. 82, no. 2, pp. 323–338, 1995. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 R. D. Meller and D. S. Kim, “The impact of preventive maintenance on system cost and buffer size,” European Journal of Operational Research, vol. 95, no. 3, pp. 577–591, 1996. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 S. M. R. Iravani and I. Duenyas, “Integrated maintenance and production control of a deteriorating production system,” IIE Transactions, vol. 34, no. 5, pp. 423–435, 2002. View at: Publisher Site  Google Scholar
 T. W. Sloan, “A periodic review production and maintenance model with random demand, deteriorating equipment, and binomial yield,” Journal of the Operational Research Society, vol. 55, no. 6, pp. 647–656, 2004. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 X. Yao, X. Xie, M. C. Fu, and S. I. Marcus, “Optimal joint preventive maintenance and production policies,” Naval Research Logistics, vol. 52, no. 7, pp. 668–681, 2005. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 N. Rezg, A. Chelbi, and X. Xie, “Modeling and optimizing a joint inventory control and preventive maintenance strategy for a randomly failing production unit: analytical and simulation approaches,” International Journal of Computer Integrated Manufacturing, vol. 18, no. 23, pp. 225–235, 2005. View at: Publisher Site  Google Scholar
 T. D. Dimitrakos and E. G. Kyriakidis, “A semiMarkov decision algorithm for the maintenance of a production system with buffer capacity and continuous repair times,” International Journal of Production Economics, vol. 111, no. 2, pp. 752–762, 2008. View at: Publisher Site  Google Scholar
 C. C. Karamatsoukis and E. G. Kyriakidis, “Optimal maintenance of a productioninventory system with idle periods,” European Journal of Operational Research, vol. 196, no. 2, pp. 744–751, 2009. View at: Publisher Site  Google Scholar  Zentralblatt MATH
 L. A. Hadidi, U. M. AlTurki, and A. Rahim, “Integrated models in production planning and scheduling, maintenance and quality: a review,” International Journal of Industrial and Systems Engineering, vol. 10, no. 1, pp. 21–50, 2012. View at: Publisher Site  Google Scholar
 C. Derman, Finite State Markovian Decision Processes, Mathematics in Science and Engineering, Academic Press, New York, NY, USA, 1970. View at: Zentralblatt MATH
 S. Ross, Introduction to Stochastic Dynamic Programming, Probability and Mathematical Statistics, Academic Press, New York, NY, USA, 1983.
 H. C. Tijms, Stochastic Models: An Algorithmic Approach, Wiley Series in Probability and Mathematical Statistics: Applied Probability and Statistics, John Wiley & Sons, Chichester, UK, 1994.
Copyright
Copyright © 2012 Constantinos C. Karamatsoukis and Epaminondas G. Kyriakidis. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.