Condition-based Maintenance Policy Research Articles

Optimization of condition-based maintenance for multi-state deterioration systems under random shock

This paper proposes a new condition-based maintenance policy for systems subject to multi-state deterioration and random shock simultaneously. Both Semi-Markov process and compound Poisson process are employed to explicitly model the exposed failure processes and dependent relationships between them. A condition-based inspection policy is designed to timely capture the states where maintenance actions are needed. The influence factors of preventive maintenance are specified as random shock and ageing process respectively. The Monte Carlo simulation technique is employed to obtain the optimal values about the proposed maintenance policy. An illustrate example about a power generator is taken into consideration to show how the newly developed maintenance policy works in real practice. The comparison analysis demonstrates that (1) the proposed condition-based inspection can effectively reveal the maintenance time points within limited inspection cost; (2) the new maintenance policy can achieve better objective values of availability and maintenance cost.

Condition-based opportunistic maintenance policies with two-phase inspections for continuous-state systems

This paper proposes two condition-based opportunistic maintenance policies with two-phase inspections for continuously degraded systems with real-time monitoring and periodic inspections, respectively. The degradation of the components and the system is modeled by a Gamma process. Existing literature finds the optimal maintenance strategies by monitoring the performance of all components in a multi-component system. However, it maybe unrealistic and costly to perform inspections on all components for certain systems. Hence, this study considers a two-phase inspection method, that is, the system-level inspection is conducted first, and then the decision for component-level inspection is made based on the system performance. Maintenance is performed to the component when its degradation reaches the preventive maintenance (PM) threshold or the corrective maintenance (CM) threshold at a component-level inspection. Whenever a maintenance action is implemented, the other components whose degradation exceeds an opportunistic maintenance (OM) threshold will be repaired. Then the long-run expected cost for two-component systems and multi-component systems are derived based on the semi-regenerative properties and simulation method. The combination of the system-level inspection interval, OM, PM and CM thresholds is determined to minimize the long-run expected cost. Finally, this study applies the proposed maintenance policies to the battery pack systems in electrical vehicles.

A predictive multi-objective condition-based maintenance (CBM) policy considering ecological factors

PurposeThe purpose of this paper is to propose an improved multi-objective optimization model for the condition-based maintenance (CBM) of single-component systems which considers periodic imperfect maintenance and ecological factors.Design/methodology/approachBased on the application of non-periodic preventive CBM, two recursion models are built for the system: hazard rate and the environmental degradation factor. This paper also established an optimal multi-objective model with a normalization process. The multiple-attribute value theory is used to obtain the optimal preventive maintenance (PM) interval. The simulation and sensitivity analyses are applied to obtain further rules.FindingsAn increase in the number of the occurrences could shorten the duration of a maintenance cycle. The maintenance techniques and maintenance efficiency could be improved by increasing system availability, reducing cost rate and improving degraded condition.Practical implicationsIn reality, a variety of environmental situations may occur subsequent to the operations of an advanced manufacturing system. This model could be applied in real cases to help the manufacturers better discover the optimal maintenance cycle with minimized cost and degraded condition of the environment, helping the corporations better fulfill their CSR as well.Originality/valuePrevious research on single-component condition-based predictive maintenance usually focused on the maintenance costs and availability of a system, while ignoring the possible pollution from system operations. This paper proposed a modified multi-objective optimization model considering environment influence which could more comprehensively analyze the factors affecting PM interval.

Optimizing a condition-based maintenance policy by taking the preferences of a risk-averse decision maker into account

Reasonably accurate remaining-useful-life (RUL) predictions allow for the introduction of maintenance policies where resources, such as spare parts and personnel, are only acquired based on the predicted need. For some assets, such a policy will help reduce the cost of renewals but will also increase the probability of renewal cycles with long downtime and associated large losses. From a decision theoretical point of view decision makers are often risk-averse and therefore their financial risk tolerance should be considered. This paper presents a procedure based on expected utility theory for the optimization challenge. To calculate the expected utility the characteristic function is used to find the full probability mass function of the maintenance cost in a finite time interval. A numerical example and a case study, based on data from an offshore oil and gas platform, are presented to illustrate the proposed model. These examples show that using the long-run cost rate to optimize the presented maintenance policy may lead to decisions that are not in line with the preferences of a risk-averse decision maker.

Optimum condition-based maintenance policy with dynamic inspections based on reinforcement learning

During the service life, inspections and repairs should be applied timely to maintain the reliability level of deteriorating structures. Condition-based maintenance (CBM) is an effective maintenance policy to reduce the life cycle cost. When the number of inspections does not change regardless of the performance, the CBM is categorize as fixed inspection (FI), otherwise, the inspection policy is denoted as dynamic inspection (DI). Compared with FI policy, DI policy performs the inspections based on the actual state and can avoid the unnecessary or insufficient inspections. Reinforcement learning is an effective and advanced decision-making tool and provides a useful method to optimize DI policy. Meanwhile, reinforcement learning has two methods (model free and model based) distinguished by the interaction method of environment. Comparison of two methods can help select an appropriate method to derive DI policy. Here, model based dynamic inspection (MBDI) and model free dynamic inspection (MFDI) are investigated for their performances in integrity management of fatigue structures. A fatigue details of ship structure is applied to illustrate the proposed framework and comparison study between DI and FI is performed. Results show that dynamic inspections can effectively reduce the expected life cycle costs. Furthermore, MFDI has a better performance than MBDI under different deteriorating rate and cost conditions.

On the explainability of convolutional neural networks processing ultrasonic guided waves for damage diagnosis

Among the maintenance policies adopted to guarantee the safety of structures throughout their service life, condition-based maintenance policies driven by structural health monitoring approaches have progressively gained importance over the last years. Within this field, among the several methods proposed in the literature to diagnose damage affecting thin-walled structures, satisfactory performances have been achieved by adopting tomographic algorithms to process ultrasonic guided waves, even though with some limitations. Recently, such limitations have been overcome by adopting machine learning-based algorithms, even though their implementation in real life applications is still limited because of the mistrust in neural networks determined by their black box-like nature. To date, however, several explainability algorithms have been proposed to interpret the behaviour of neural networks, in particular in the medical and in the military fields, where trust in the tools adopted must be guaranteed. Thus, exploiting the potentialities of explainability frameworks, in this work the layer-wise relevance propagation algorithm is employed to explain the predictions of convolutional neural networks for classification and for regression that characterise damage by processing ultrasonic guided waves excited and sensed by means of a 2-D network of piezoelectric devices. First, the explainability algorithm is applied to give a score to each sample of the acquired ultrasonic guided waves, then such scores are collected by means of a properly developed aggregation strategy to rank the most informative couples of piezoelectric devices. The capabilities of the explainable damage diagnosis framework are demonstrated by means of a numerical, yet realistic, case study involving a metal plate affected by crack-like damage. In particular, the focus is set on the explanation of the behaviour of the neural networks involved, with the aim of building trust in such algorithms and, possibly, revealing damage-related hidden features of ultrasonic guided waves.

Online Improvement of Condition-Baesd Maintenance Policy via Monte Carlo Tree Search.

This article proposes a method of scheduling maintenance in complex manufacturing systems in scenarios where there is frequent competition for maintenance resources. We use a condition-based maintenance policy that prescribes maintenance actions based on a machine's current health. However, when several machines are due for maintenance, a maintenance technician must choose between multiple competing jobs. While a common approach is to establish rules that dictate how maintenance jobs should be prioritized, such as the first-in, first-out rule, the goal of this work is to improve upon static policies in real time. We do this by strategically evaluating sequences of maintenance actions and playing out many "what-if" scenarios to see how the system will behave in the future. Implementation of the proposed method relies on the construction of a simulation model of the target system. This model is capable of retrieving the current state of the physical system, including the degradation state of machines, the availability of maintenance resources, and the distribution of parts throughout buffers in the system. We present several simulation experiments that demonstrate the improvement in system performance that our approach provides. Future work will aim to improve the efficiency of maintenance prioritization through online learning as well as more accurately identify manufacturing system configurations that will yield the greatest benefit of these methods.

철도차량의 좌우측절단자료를 이용한 2단계 상태기반 점검주기 연구

Purpose: This paper introduces two-phase condition-based maintenance by analyzing left-truncated and right-censored field data, with characteristics managed by the existing maintenance schedule.BRMethods: The failure rate was initially observed as a constant failure rate (CFR) and, after a certain point, as an increasing failure rate (IFR). Therefore, a two-phase condition-based maintenance policy with each stage maintenance cycle was proposed. In addition, a maintenance model was developed to determine the optimal maintenance schedule in terms of economic feasibility.BRResults: The optimal maintenance schedule was derived by minimizing the total cost, including the inspection cost, failure cost, and improvement factor. Further, changes in the optimal maintenance schedule were examined based on the ratios of the inspection cost, failure cost, and improvement factor.BRConclusion: Designing a maintenance policy to evaluate the system conditions and mutual adequacy between the cost and improvement effect in terms of light and medium maintenance is essential.

Joint optimization of condition-based performance control and maintenance policies for mission-critical systems

The failure behavior of mission-critical systems typically depends on their performance levels, which provides opportunities to control deterioration by adjusting the performance levels. In addition to performance level selection, preventive maintenance decision is another factor affecting system survivability and mission success probability, providing a new train of thought for joint optimization of preventive maintenance and performance control for effective risk control of mission-critical systems. In existing research, mission-based maintenance and performance control decisions are seldom considered. This paper focuses on mission-critical systems and investigates the joint optimization of maintenance and performance control policies to balance mission success probability and system survivability based on the workload and system state. The problem is formulated as a Markov decision process and structural properties are analyzed to determine the optimal maintenance and performance control policies. The maintenance strategies exhibit a threshold structure, which strongly depends on the relationship between the performance level and the degradation process. For the purpose of comparison, the performance of several heuristic strategies is analytically evaluated. The example of the data transmission satellite network is given to demonstrate the superiority of the proposed strategies in reducing operations costs.

Condition-based maintenance strategy for redundant systems with arbitrary structures using improved reinforcement learning

The condition-based maintenance (CBM) decision-making for redundant systems has attracted increasing attention. Most existing studies are dedicated to k-out-of-n redundant systems and the search of the optimal maintenance policy is efficient for low-dimensional CBM. In practical applications, complex system structures and failure criteria are commonly observed, posing challenges for searching the optimal CBM policy. This paper studies the optimal CBM strategy for redundant systems with arbitrary system structures using improved reinforcement learning, considering failure and economic dependences. The decisions of imperfect repair and replacement of failed components are considered dynamically, and an efficient solution method of dynamic maintenance strategy is investigated via improved reinforcement learning incorporating re-learning and pre-learning processes. Numerical studies are conducted and the results indicate that the proposed method is effective in reducing the maintenance cost and efficient in searching the optimal CBM strategy for redundant systems.

A general model for life-cycle cost analysis of Condition-Based Maintenance enabled by PHM capabilities

In this work, we propose a general modelling approach to estimate the life cycle cost of a system equipped with Prognostics and Health Management (PHM) capabilities, undergoing a Condition-Based Maintenance (CBM) policy. The approach builds on the Markov Chain theoretical framework, with transition probabilities linked to both PHM performance metrics of the literature and a novel metric. The developed approach can be used to guide economic decisions about CBM development, whichever the PHM algorithm is but provided that its performance metrics are estimated. The model is validated through a case study concerning a mechanical component of a train bogie affected by fatigue degradation, considering two different prognostic algorithms: Particle Filtering and a Model-Based approach.

Threshold policy of condition-based maintenance for a multi-state system and deteriorating sensor

Threshold policy of condition-based maintenance for a multi-state system and deteriorating sensor

A condition-based maintenance policy for continuously monitored multi-component systems with economic and stochastic dependence

In this paper we propose a condition-based maintenance policy for continuously monitored multi-component systems subject to stochastic dependence through load sharing and economic dependence through maintenance set-up costs. Stochastic dependence provides an incentive to replace failed components as soon as possible, since component failures increase the load, and hence the deterioration rates, of the remaining components. In contrast, economic dependence encourages the clustering of component replacements to reduce maintenance frequency, resulting in less downtime and fewer maintenance set-up costs. In this paper, we propose a novel, condition-based maintenance policy to obtain the optimal replacement decisions at maintenance opportunities. Through numerical studies we see the importance of a policy that incorporates both types of system dependence. The policy incorporates a utility/reward function that is a combination of interpretable penalties that encapsulate the costs of stochastic and economic dependence. The utility function trades off the rewards of clustering components with the loss due to load sharing. The policy minimizes the overall cost of the system by choosing actions that minimize the total long-term penalty. We show that the proposed policy outperforms various alternative policies by reducing system life-cycle costs.

Condition-Based Monitoring and Maintenance: State of the Art Review

Manufacturing firms face great pressure to reduce downtime as well as maintenance costs. Condition-based maintenance (CBM) can be used to effectively manage operations and maintenance by monitoring detailed machine health information. CBM policies and the development of the mathematical models have been growing recently. This paper provides a review of the theoretical and practical development in the field of condition-based maintenance and its current advancements. Standard CBM platform could make it effective and efficient in implementation and performance improvement.

An Integrated Optimization of Lot-Sizing and Condition-Based Maintenance Policy of a Two-Stage Deteriorating System

An Integrated Optimization of Lot-Sizing and Condition-Based Maintenance Policy of a Two-Stage Deteriorating System

Joint optimization of condition-based maintenance policy and buffer capacity for a two-unit series system

The purpose of this paper is to address a joint optimization problem of condition-based maintenance policy and buffer capacity for a two-unit series system. The system consists of two machines and an intermediate buffer, both upstream and downstream machines suffer from soft failure, and maintenance actions are taken according to the deterioration state of the system. Different from previous studies that the maintenance strategy and buffer capacity were always be optimized separately, this paper optimizes the maintenance strategy as well as the buffer capacity. All system states and transition probabilities between them are analyzed based on semi-Markov process, and then the expected cost and sojourn time of the system are constructed. By minimizing the average operation and maintenance cost per unit time of the system, joint optimization decision (i.e., optimal maintenance thresholds of upstream and downstream machines and buffer size) were determined. A practical simulation example of production-buffer-assembly system is used to illustrate the superiority of the proposed joint optimization strategy.

A condition-based maintenance policy for multi-component systems subject to stochastic and economic dependencies

Condition-based maintenance (CBM) optimization utilizing prognostic information and methods for complex systems has attracted increasing attention. This paper presents a bi-level approach to optimize a CBM policy for multi-component systems subject to stochastic and economic dependencies. The objective of the decision rule at the system level is to address if preventive and/or corrective maintenance actions are necessary regarding the predictive reliability of the system. At the component level, the aim is to identify the optimal group of components to be preventively maintained when maintenance is triggered due to the decision made at the system level. A maintenance efficacy indicator is introduced to prioritize components to be maintained by taking into account both the resulting reliability improvement quantity and the maintenance cost incurred. The optimal inter-inspection interval, the thresholds of system and component conditional reliability for triggering maintenance are jointly determined for minimization of the long-run average cost rate. A numerical study and sensitivity analysis on a 2oo3:G system are carried out to validate the performance of the proposed maintenance policy. Compared to another opportunistic maintenance policy without considering the system health condition, the proposed CBM policy leads to more cost gain and effectiveness in the maintenance decision-making.

A Bayesian condition-based maintenance and monitoring policy with variable sampling intervals

The aim of this study is to propose a Condition-Based Maintenance and Monitoring (CBM) policy which employs a Bayesian inspection scheme. A single unit system, which is subject to both operational deterioration and catastrophic failures, is considered. The equipment may operate in two different non-observable states (healthy and unhealthy). The unhealthy state is characterized by higher operational cost and higher proneness to failure. Failures are self-announced (directly observable) and thus, corrective maintenance is implemented immediately. A new double-sampling Bayesian control chart with state-dependent variable inspection frequency is proposed. The process operation is analytically modeled through a six-state Markov process, while, unlike all previous Bayesian models, there is no need for discretization of the unhealthy-state probabilities. At each inspection instance all available information regarding the equipment condition is utilized in order to schedule future inspections and preventive maintenance actions and detect possible operation in the unhealthy state. The critical parameters, namely the duration of the inspection intervals, the sample sizes and the preventive maintenance times, which minimize the expected total cost per time unit, are determined. Numerical comparisons with three other Bayesian CBM models are conducted to demonstrate the effectiveness of the proposed policy.

Design of a Condition-based Maintenance Policy Using a Surrogate Variable

Design of a Condition-based Maintenance Policy Using a Surrogate Variable

Condition-based maintenance strategies for stochastically dependent systems using Nested Lévy copulas

Nowadays, industrial systems are more and more complex and components are integrated dependently. Without consideration of dependence between components often leads to overestimate the reliability of system and high maintenance cost. In this paper, a multi-component system with hierarchical stochastic dependencies is considered. Components within subsystem are strongly correlated and dependence between subsystems is heterogeneous. Thanks to its capacity of modeling asymmetric dependence structures, the Nested Lévy copula is used in this paper. Based on the multiple sources of dependence among components and hierarchical structures, a novel condition-based maintenance policy is proposed. The inspection scheme as well as maintenance decision is dynamically planned according to the collected inspection information and the influence of stochastic dependence is investigated. The numerical results show the efficiency and robustness of proposed maintenance policy on reducing the inspection and maintenance cost.