Optimization of maintenance policy under parameter uncertainty using portfolio theory

Burn-in is a widely used method to eliminate the initial failures. Preventive maintenance policy such as age replacement is often used in field operation. In this paper burn-in and maintenance policy are taken into consideration at the same time. The properties of the corresponding optimal burn-in times and optimal maintenance policy are discussed.

Thompson1 considers a linear control model to ascertain the optimal preventive maintenance policy for production equipment and, simultaneously, its optimal sale date. He assumes age‐dependent deterministic continuous deterioration, production rate and linear efficiency of maintenance expenditures. In the present paper we first consider a linear maintenance model under a budget constraint. The impact of such a restriction is an earlier switch in the optimal bang‐bang policy. Then a non‐linear, non‐autonomous generalization of Thompson's model is analysed by the maximum principle. The optimal maintenance policy is shown to be downward sloping provided that in each time period the production rate is greater than the sum of interest and deterioration rate. The monotonicity of the optimal rate of preventive maintenance holds also under the additional possibility of random machine failure provided the production rate is greater than the sum of the rate of interest, the rate of deterioration and the rate of stochastic failure at each age. Moreover, it is shown that the higher the failure rate the smaller the optimal expenditures for preventive maintenance at any fixed age of the machine. In particular, the optimal maintenance rate at age t in a model without random failure is greater than those in any case including stochastic failure at the same age.

Different type of maintenance problems have been investigated during the past decades. With the growth in complexity of modern systems, the cost of purchasing new systems becomes higher ever than before. Surprisingly, very little attention in the literature has been paid to the failure replacement in these preventive maintenance models. However, in practice most systems incur increasingly expensive down time costs when the systems suffer the unexpected failures. The major changes to the standard maintenance model in this research are the addition of a failure replacement cost for a failed system and the reliability limit for which system has to be replaced. The overall objective of this research is to propose an efficient algorithm to obtain the optimal maintenance policy. The optimal preventive maintenance policy is investigated with incorporating the costs of minimal repair, imperfect preventive maintenance, failure replacement and preventive replacement. The specific maintenance actions for a repairable system are when to perform an imperfect preventive maintenance or to replace the system while the system is in the operating condition. In another case, the specific maintenance actions for a repairable system are whether to replace the system, or to perform a minimal repair when the system fails. The number of imperfect preventive maintenance and the control limits of the optimal maintenance policy are calculated under the expected average cost rate criterion.

Optimal maintenance policy and residual life estimation for a slowly degrading system subject to condition monitoring

This paper proposes optimal maintenance policies for multi-deteriorating product considering its remanufacturability so as to reduce maintenance cost. It is assumed that all the failures can be repaired and the repair time can be considered. A maintenance model is established for diverse health index based on long-run average cost rate. In the model, discrete Markov process is adopted to depict the system deteriorating. Once a failure occurs, the system's health index can be identified immediately. Then the system's remanufacturability and continuing using capability through repairs are evaluated to determine the maintenance policy. The optimal maintenance policies can be obtained by a policy-iteration algorithm. Finally, the proposed approach is applied to an automotive engine. The results show that the model can describe the multi-degradation repair process effectively, and can be used to optimize the maintenance policies.

This paper investigates the optimal threshold values of age to perform preventive maintenance (PM) actions for leased equipment within the lease period. In this paper, we use age reduction method to describe the degree of PM and construct the maintenance cost function. For repairable leased equipment, two maintenance models are proposed: (i) maintenance policy of single-phase and (ii) maintenance policy of two-phase. During the lease period, PM actions are carried out when the age of equipment reaches a certain threshold value. Any failure of the leased equipment is rectified by a minimal repair within the lease period. Under these maintenance schemes, the mathematical models of the expected total cost for maintenance policies of single-phase and two-phase are established, and the optimal maintenance policies are derived such that the expected total cost is minimized. Finally, the features of the optimal maintenance policy are illustrated through numerical examples.

Joint optimization of maintenance and spares inventory policy for a series-parallel system considering dependent failure processes

Operating a system for optimum availability is a key aspect in many a complex system. Failure of the system can have serious consequences both in monetary terms as well as in loss of production or disruption of service. Optimal control theory provides a convenient tool to obtain the optimal maintenance policies of maintainable systems. These optimal policies are obtained by applying the generalized maximum principle of Gamkrelidze. The prescribed objective function is represented in terms of the cost of repair as well as the downtime cost. The models considered are Markov in that their present state provides sufficient information about the system. Constant as well as time-varying optimal maintenance policies during the planning or operating period can be considered by the proposed method. The optimal maintenance policy very much depends on the failure rates of the individual system components.

The semiconductor industry is rapidly evolving and is foreseen to grow at a fast pace. As this global world is technology-based with high dependency on machine operations, the performance of machines in meeting production demand is crucial in determining the survival of the industry. Therefore, an optimal maintenance policy must be determined in order for these machines to maximise their performance. Nevertheless, maintenance policy evaluation process has always been complicated as it involves different kinds of maintenance policies, as well as industry objectives and requirements. This paper presents a systematic maintenance policy evaluation model to assist the industrial management in performing analysis and making decision using the technique for order preference by similarity to ideal solutions (TOPSIS). The case study conducted demonstrated the viability and significance of the proposed model.

This article considers an extended parallel system with cascading failure: Compared with the conventional parallel system with independent failure, the factor that cascading failure could reduce the reliability of the system due to redistribution of the workload. Reliability measures of the system are derived, using the theory of stochastic processes. To counteract the impact of cascading failure, a preventive maintenance policy is proposed. Further, this is applied to a dual active system. An optimal maintenance policy which minimizes the expected cost rate is derived. Finally, a numerical example is given and some discussions are made.

Burn-in is a widely used method to improve quality of products after they have been produced. For a repairable component there are two common types of repair, complete repair and minimal repair. Preventive maintenance policies such as age replacement and block replacement are often employed in field operation. The present paper takes burn-in, maintenance and repair into consideration at the same time and considers related cost structures. The properties of the corresponding optimal burn-in times and optimal maintenance policies are discussed.

Burn-in is a widely used method to improve quality of products after they have been produced. For a repairable component there are two common types of repair, complete repair and minimal repair. Preventive maintenance policies such as age replacement and block replacement are often employed in field operation. The present paper takes burn-in, maintenance and repair into consideration at the same time and considers related cost structures. The properties of the corresponding optimal burn-in times and optimal maintenance policies are discussed.

Optimal maintenance policy for a system subject to damage in a discrete time process

Optimal preventive maintenance policy for leased equipment using failure rate reduction

PurposeThe purpose of the current work is to present an intuitionistic fuzzy (IF) concept-based structured framework for selecting the optimal maintenance policy in a sugar mill.Design/methodology/approachThe proposed framework utilizes IF concept based multi-criteria decision making (MCDM) approaches, specifically IF-analytic hierarchy process (IF-AHP) and IF-Technique for Order of Preference by Similarity to Ideal Solution (IF-TOPSIS). IF theory based MCDM approaches utilises the hesitation present in the maintenance experts to encounter the uncertainties/vagueness to much higher degree of accuracy in decision-making. The suggested framework is used to evaluate and select the optimal maintenance policy based on six different criteria namely safety factor, cost factor, maintenance factor, reliability, risk and added values.FindingsThe IF-AHP approach has been employed to calculate the weights of the criteria and sub-criteria, while the IF-TOPSIS approach was utilized to rank the maintenance strategies. Based on the results, with a relative coefficient value of 0.7204, corrective maintenance (CM) is determined to be the best maintenance policy. For certifying the consistency of the recommended structured framework, sensitivity analysis (SA) has been also conducted.Research limitations/implicationsThe ranking results obtained from the analysis are provided to the maintenance management of the considered sugar mill for its further implementation and validation. The findings of this work are also applicable to all other sugar mill industries which are installed globally.Practical implicationsThe analysis results has been supplied to the maintenance manager of considered sugar mill industry. The implementation policy of the results will be shared with the higher management and hence once implemented the results could be tested and verified.Originality/valueThe developed framework so implemented to the considered sugar mill industry is original in nature. Also, consideration of hesitation effect in the collected raw data under the developed framework provide more authenticated decision results which proves to be useful in achieving higher availability and profitability of the industry.