Degradation prediction and rolling predictive maintenance policy for multi-sensor systems based on two-dimensional self-attention

Traditional preventive maintenance policy gradually failed to guarantee the security and economy of current mechanical systems. This paper proposed a highly efficient rolling predictive maintenance (RPdM) policy for multi-sensor system, to make maintenance decisions. In this policy, to cope with the uncertainty of remaining useful life (RUL) prediction, the degradation process of the system is first divided into four intervals according to the inspection interval and spare parts lead time. Then, the two-dimensional self-attention (TDSA) method, which extract time dimensional and feature dimensional features by parallel computation, is developed to predict the probabilities of system RUL in the four intervals instead of the point of RUL. In addition, the output probabilities of the TDSA method are utilized to calculate the maintenance cost rate of the current inspection point and future point. The maintenance decision including spare parts ordering time and maintenance time is determined by comparing the maintenance cost rate of each inspection point, and the decision is updated at the next inspection point. To verify the effectiveness of the proposed RPdM policy, the C-MAPSS dataset provided by NASA is employed to implement degradation prediction and maintenance decision. Experiment results show that the cost rate of RPdM policy is lower than preventive maintenance policy, and only 27.7% higher than ideal maintenance policy which is impossible in real engineering. Moreover, the impact of different out-of-stock costs and corrective costs are explored and shows the good robustness of the RPdM policy.     

Optimally maintaining a multi-state system with limited imperfect preventive repairs

In this paper, the optimal maintenance policy for a multi-state system with no observation is considered. Different from most existing works, only a limited number of imperfect preventive maintenance actions can be performed between two successive replacements. Assume that the system's deterioration state cannot be observed during its operation expected after each replacement, and it evolves as a discrete-time Markov chain with a finite state space. After choosing the information state as state variable, the problem is then formulated as a Markov decision process over the infinite time horizon. In order to increase the computational efficiency, several key structural properties are developed by minimising the total expected cost per unit time. The existence of the optimal threshold-type maintenance policy is proved and the monotonicity of the threshold is obtained. Finally, a numerical example is given to illustrate the optimal policy.     

Optimal inspection and economical production quantity strategy for an imperfect production process

This paper considers a problem relating to the integration of production, maintenance, inspection and inventory. The optimal inspection interval, inspection frequency and production quantity were studied, assuming an imperfect production process and a maintenance policy that permits minimal repair, shortages and inspection errors. When the process is in the in-control state, preventive maintenance is performed. If the process is in the out-of-control state, there are two possibilities. The first is a relatively mild production process shift, where minimal repair can bring the process back into the in-control state. The second possibility is a more serious shift, where minimal repair cannot bring the process back into the in-control state. In this case, it will be necessary to halt production and perform repair or replacement. It is accepted that, when a production process is in the out-of-control state, an amount of defective goods will be produced. Within the context of a deteriorating production process system, the advantages of preventive maintenance were explored, using numerical analysis to analyse the impact of inspection errors, permitted shortage conditions and minimal repair on cost.     

Two-stage Maintenance of a Production System with Exponentially Distributed On- and Off-periods

We consider an on–off production system which is subject to failure during on-periods. In case of a failure, the production system is maintained correctively. In addition, preventive maintenance is carried out to prevent failures. The costs of both preventive and corrective maintenance are modelled in terms of the net down-time of the production system, that is, the time that the system is not available when needed for production. Since preventive maintenance can be planned during off-periods, there is a perspective of significant savings if some freedom is built in concerning the starting time of preventive maintenance. Therefore, a two-stage maintenance policy is considered, which—in a first stage—provides the maintenance manager with a finite interval during which preventive maintenance must be carried out, and—in a second stage—determines the optimal starting time for preventive maintenance within this interval. Computational results offer useful insights, and indicate that significant savings can be achieved in comparison with a classical age maintenance policy.     

A condition-based maintenance strategy for heterogeneous populations

This paper develops a maintenance strategy, called inspection–replacement policy, to cope with heterogeneous populations. Burn-in is the procedure by which most of the defective products in a heterogeneous population can be identified and removed prior to being placed in service. However, modern manufacturing is so well developed that a defective product is able to function for a long period of time even under aggravated operational conditions. Instead of weeding defective products out via costly burn-in tests, use can be made of them in field operation where maintaining actions will be performed to prevent early in-use failures. The inspection–replacement policy consists of an inspection, conducted in an early stage with the purpose of identifying and replacing defective products, and a preventive replacement, carried out at a later stage to prevent wear-out failures. The preventive-replacement time is dynamically determined, depending on the information obtained by the inspection. The inspection–replacement policy is compared with a joint burn-in and age-based-replacement policy to show its practicability and competence.     

两机器串联生产系统在随机状态下的最优生产和维护策略的研究

对机器状态的变化为非齐次马尔可夫过程,带内部缓冲栈的单工件－两机器串联生产系统的最优生产率和维护率进行研究,给出了最优维护率的结构,得到最优生产率由安全面和临界面确定的结构特性,针对解析解无法获得,在得出的结构特性的基础上,给出了一种启发式的次优控制策略。     

基于可用度分析的故障检查间隔期的确定

为了解决由于检查和维修引起的停机时间不能忽略的情况,文章分析了产品的多种检查维修的情况,并应用预防性维修理论,建立了故障检查模型。通过应用此故障检查模型,期望得到最大可用度,以及在此可用度下的维修检查间隔期,从而达到预计的故障检查目标,实现对故障检查间隔期的优化。通过算例分析,在求得最大可用度的情况下,提出了可行的故障检查间隔期,证明了该模型的实用性和有效性。     

基于故障率修正参数的生产设备预防性维修研究

点检定修制是生产设备预防性维修的1种重要方法,是全员、全过程对设备进行动态管理的1种设备管理方法。点检定修制的核心是确定点检周期,即预防性维修周期。为此引入故障率修正参数,建立了基于单位时间净生产效益的生产设备预防性维修模型,并通过仿真的方法确立最佳预防性维修周期和维修次数,阐述了维修程度的不同对预防性维修周期的影响以及预防性维修对企业增加利润的重要性。结果表明,选择合理的故障率修正参数可以有效降低故障率,增大平均单位产值,从而为管理者制定有效的维修计划提供决策。     

A preventive maintenance policy for a standby system subject to internal failures and external shocks with loss of units

In many situations, serious damage and considerable financial losses are caused by non-repairable failures of a system. Redundant systems and maintenance policies are commonly employed to improve reliability. This paper is focused on the modelling of a complex cold standby system by analysing the effectiveness and costs of preventive maintenance, always in an algorithmic form. The online unit of the system is subject to wear failures and external shocks. The online unit can go through an indeterminate number of degradation levels before failure. This one is observed when inspections occur. Inspections are performed at random intervals, and when one takes place, the unit is taken to the preventive maintenance facility if it is necessary. The preventive maintenance time and cost is different depending on the degradation level observed. If only one unit is performing, a minimal maintenance policy is adopted in order to optimise system behaviour. Reliability measures such as the conditional probability of failure are worked out in a well-structured and algebraic form in transient and stationary regimes by using algorithmic methods. The stationary distribution is calculated using matrix analytic methods, and rewards are included in the model. An optimisation example shows the versatility of the model presented.     

Risk based opportunistic maintenance model for complex mechanical systems

The essence of the complex mechanical system can be considered as an open system. There exist coupling relationships between various parts in the system and also between different fault modes, which result in multiple fault propagation paths. Considering the safety, benefits and maintenance loss, parameters such as downtime losses, minimal repair costs, corrective, preventive and opportunistic maintenance costs, should be analyzed comprehensively to investigate the influence of different maintenance strategies.A new risk based opportunistic maintenance (RBOM) model considering failure risk is proposed in the paper. It helps to convert the negative random factors caused by single faults to a favorable opportunity of preventive defense against failure for other slight degraded components in advance, so the overall economic losses could be reduced. The global optimization algorithm is further developed to realize RBOM policy. Case studies are provided to illustrate the proposed approaches, with sensitivity analysis of the position, time, style and criterion of the RBOM strategy. Comparative study with the widely used maintenance policy demonstrates the advantage of the proposed method can significantly reduce the maintenance cost and failure risk, and are expected to bring immediate benefits to the energy industry.     

存在故障相关及不完备检测的主辅并联系统可靠性建模与维修策略

主辅部件并联普遍存在于机电系统中, 部件间存在故障相关性等因素使得针对系统的可靠性建模和维修策略优化尤为困难. 本文以部件间存在I类和III类故障相关性及因检测系统的不完备引起主部件累积损伤速率分布改变的主辅并联系统为对象, 通过对系统退化过程的分析建立其可靠性模型. 依据所建模型和观测的辅助部件的故障次数及系统的运行时间, 建立(T, N)预防性更换策略优化模型. 通过案例分析检测系统的完备性、累积损伤速率等参数对系统可靠性及预防性维修策略的影响.     

基于服务性能合同模式下单部件系统视情维修策略研究

针对供应商系统维修的低效率以及维修成本参数较难获得的问题,提出了基于服务性能合同模式(PBC)下的单部件系统最优视情维修策略模型。首先,基于Gamma分布,描述单部件系统连续递增的退化过程,依据系统实时检测状态与预防维修阈值、故障阈值之间的关系,实施不同的维修策略;其次,分析单位更新周期内的检测次数和使故障设备恢复如新的维修方式,以供应商利润率最大化为目标函数,以最佳维修阈值与检测间隔时间为决策变量,建立以利润为中心的视情维修优化模型;最后,利用改进灰狼算法求解数学模型,通过算例验证所提出模型的有效性,并进行了各维修费用参数对目标函数以及最优维修策略的灵敏度分析。     

Preventive maintenance scheduling for repairable system with deterioration

Maintenance as an important part in manufacturing system can keep equipment in good condition. Many maintenance policies help to decrease the unexpected failures and reduce high operational cost such as conventional preventive maintenance. But these conventional preventive maintenance policies have the same time interval T that may easily neglect system’s reliability, because the system deteriorates with increased usage and age. Hence, this study has developed a reliability-centred sequential preventive maintenance model for monitored repairable deteriorating system. It is supposed that system’s reliability could be monitored continuously and perfectly, whenever it reaches the threshold R, the imperfect repair must be performed to restore the system. In this model, system’s failure rate function and operational cost are both considered by the effect of system’s corresponding condition, which helps to decide the optimal reliability threshold R and preventive maintenance cycle number. Finally, through case study, the simulation results show that the improved sequential preventive maintenance policy is more practical and efficient.     

Joint quality control and preventive maintenance strategy for imperfect production processes

In this paper, we develop a joint quality control and preventive maintenance policy for a production system producing conforming and nonconforming units. The considered system consists of one machine which must supply another production line operating on a just-in-time basis. Each lot produced by the machine is subject to a quality control. According to the proportion l of nonconforming units observed and compared to a threshold value l _m , one decides to undertake or not maintenance actions on the system. In order to palliate perturbations caused by the stopping of the machine for preventive and corrective maintenance actions of random durations, a buffer stock h is built up to ensure the continuous supply of the subsequent production line. The proposed strategy is modelled using simulation and experimental design. This approach allows to generate a second order response surface allowing to easily determine the optimal rate, l_m^*{l_{m}^*}, of nonconforming units on the basis of which preventive maintenance actions should be performed, and the optimal size, h*, of the buffer stock to be built. These values minimize the total cost per time unit which includes the costs related to maintenance, quality and inventory.     

Planning and optimizing the maintenance of paper production systems in a paper plant

Manufacturing and production plants operate physical assets that deteriorate with usage and time, thus, maintenance actions are required to restore the assets back to their original predetermined operational conditions. But since, organizational resources are limited and scarce. The objective of an effective maintenance program is to minimize total cost of inspection and repair, and equipment downtime. In this paper, we describe a new multi-criteria optimization framework for deriving optimal maintenance schedules for preventive maintenance which considers availability, maintenance cost and life cycle costs as the criteria for optimization. The Simulink toolbox of the Matlab has been interfaced with Genetic algorithm for optimization. After getting solution, Stochastic Petri nets has been used to model the system and find out the effect of optimized maintenance schedules on system performance. The application of the proposed framework has been discussed on a practical case of a paper plant.     

Periodic and sequential preventive maintenance policies over a finite planning horizon with a dynamic failure law

The purpose of this study is to propose and model periodic and sequential preventive maintenance policies for a system that performs various missions over a finite planning horizon. Each mission can have different characteristics that depend on operational and environmental conditions. These proposed preventive maintenance policies are defined and modeled mathematically. The study of these two policies is based on a dynamic system failure law that takes into account the different missions performed. The first step is to determine the optimal business plan to achieve, i.e. the set of missions to perform in order to maximize the profit of missions minus maintenance costs. Thus, for each plan, we determine the maintenance planning considering two policies. The first preventive maintenance policy is periodic and the objective is to determine the optimal number of preventive maintenance to achieve. For the second policy, namely sequential, we calculate the optimal number of preventive maintenance intervals and the duration of these different intervals.     

Preventive maintenance scheduling for a multi-component system with non-negligible replacement time

When maintenance models are developed for complex continuous operating systems, with large production losses, it is important to take into account the maintenance duration, in order to provide an effective maintenance program. In this work, a preventive maintenance model is presented in order to coordinate the component replacements in a multi-component system. The model is based on the partial periodic renewal policy. At each partial renewal, a group of components to be replaced is defined. When all components are replaced simultaneously, we consider that the system undergoes an overhaul which resets it to the new state. Moreover, the model takes into account for non-negligible replacement times, where different assumptions are considered. The relevant effect of maintenance times on the optimal policy is clearly shown by the numerical results, provided by the application in an oil refinery.     

An optimal maintenance policy based on generalized stochastic Petri nets and periodic inspection

Periodic maintenance of equipment is essential for its optimum performance, thereby enabling production efficiency. In the past, studies on preventive maintenance of automated manufacturing systems (AMS) determined the optimal preventive maintenance policy under different performance indexes. Generally, most hypotheses indicate that equipment reliability can be restored to 1.0 through preventive and corrective maintenance. However, in practical application, the implementation of preventive maintenance results in partial deterioration of equipment; moreover, the reliability of equipment cannot be restored to as‐good‐as‐new. In addition, the greater the complexity of connections of the equipment, the greater is the difficulty in determining the timing for preventive maintenance. On account of these characteristics, generalized stochastic Petri nets (GSPN) are well‐suited for the implementation of preventive maintenance. Therefore, this paper applies GSPN for deciding the optimal maintenance policy and constructing models for different levels of maintenance and renewal for an AMS with a serial‐parallel layout. As a result of the application of GSPN, the following optimal maintenance policy for an AMS was obtained in this study: Preventive maintenance conducted at intervals of every 240 hours will reduce cost by 46% as opposed to the practice of replacing defective parts when necessary. Copyright © 2010 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society     

Buffer allocation and preventive maintenance optimization in unreliable production lines

In this paper, we consider a serial production line consisting of \(n\) unreliable machines with \(n-1\) buffers. The objective is to determine the optimal preventive maintenance policy and the optimal buffer allocation that will minimize the total system cost subject to a given system throughput level. We assume that the mean time between failure of all machines will be increased after performing periodic preventive maintenance. An analytical decomposition-type approximation is used to estimate the production line throughput. The optimal design problem is formulated as a combinatorial optimization one where the decision variables are buffer levels and times between preventive maintenance. To solve this problem, the extended great deluge algorithm is proposed. Illustrative numerical examples are presented to illustrate the model.     

Preventive-maintenance policy for leased products under various maintenance costs

This paper investigates the effects of preventive-maintenance cost functions on the optimal preventive-maintenance policy for a leased product with Weibull life-time distribution. During the lease period, any product failures are rectified by minimal repairs and may incur a penalty to the lessor, if the time duration for performing a minimal repair exceeds a pre-specified time limit. To reduce repair costs and possible penalty, preventive-maintenance actions are scheduled in the lease contract. The objective of this paper is to derive the optimal preventive-maintenance schedule and maintenance degrees such that the expected total cost is minimized. Some structural properties of the optimal policy are obtained and an efficient algorithm is provided to search for the optimal policy. For the cases where the preventive maintenance cost is a constant or a linearly increasing function, the effects of the preventive-maintenance cost function on the optimal policy are investigated in detail both theoretically and numerically.