Predictive Maintenance Policy Research Articles

A predictive maintenance policy for manufacturing systems considering degradation of health monitoring device

Health monitoring device (HMD) providing the current condition for its monitored manufacturing system is often subjected to a gradual degradation process. The degradation of the HMD may lead to inaccurate system condition monitoring data, conducting to non-appropriate maintenance decision-making. This paper presents a novel predictive maintenance policy for manufacturing systems considering degradation of its HMD. To address this issue, a method based on Kalman filter is developed for estimating the system state. However, due to the data uncertainty given by statistic noise and HMD degradation impacts, the error between estimated value and true state is unavoidable. To overcome this issue, an adaptive predictive maintenance, in which both the estimated state of system and the current state of HMD are used for maintenance decision-making, is then proposed for the system and its HMD. A cost model is developed to find the optimal maintenance policy. A case study of injection molding machine is conducted to show the applications and the benefits of the proposed maintenance policy.

Degradation modeling and opportunistic maintenance for two-component systems with an intermittent operation component

In this study, an opportunistic predictive maintenance framework was developed for a system comprising one continuous operation component (COC) and intermittent operation component (IOC). The COC operates consistently with the system. In contrast, the IOC operates alternately in operation and standby modes. The degradation of the COC can increase the working frequency of the IOC, which can accelerate its degradation process. Omitting this degradation interaction in degradation modeling and reliability assessment could result in inaccurate lifetime prognostic and suboptimal maintenance plans. Therefore, in this study, a degradation model for the IOC is proposed considering the influence of the degradation state of the COC. A reliability model for the IOC was derived and used for predictive maintenance. Based on the proposed degradation models, an opportunistic predictive maintenance policy was proposed, considering the structural and economic dependencies between the components. Finally, a case study of an air compressor system is conducted to demonstrate the viability and benefits of the proposed model.

A metric for assessing and optimizing data-driven prognostic algorithms for predictive maintenance

Prognostic Health Management aims to predict the Remaining Useful Life (RUL) of degrading components/systems utilizing monitoring data. These RUL predictions form the basis for optimizing maintenance planning in a Predictive Maintenance (PdM) paradigm. We here propose a metric for assessing data-driven prognostic algorithms based on their impact on downstream PdM decisions. The metric is defined in association with a decision setting and a corresponding PdM policy. We consider two typical PdM decision settings, namely component ordering and/or replacement planning, for which we investigate and improve PdM policies that are commonly utilized in the literature. All policies are evaluated via the data-based estimation of the long-run expected maintenance cost per unit time, using monitored run-to-failure experiments. The policy evaluation enables the estimation of the proposed metric. We employ the metric as an objective function for optimizing heuristic PdM policies and algorithms’ hyperparameters. The effect of different PdM policies on the metric is initially investigated through a theoretical numerical example. Subsequently, we employ four data-driven prognostic algorithms on a simulated turbofan engine degradation problem, and investigate the joint effect of prognostic algorithm and PdM policy on the metric, resulting in a decision-oriented performance assessment of these algorithms.

Data-Driven Predictive Maintenance Policy Based on Dynamic Probability Distribution Prediction of Remaining Useful Life

As the reliability, availability, maintainability, and safety of industrial equipment have become crucial in the context of intelligent manufacturing, there are increasing expectations and requirements for maintenance policies. Compared with traditional methods, data-driven Predictive Maintenance (PdM), a superior approach to equipment and system maintenance, has been paid considerable attention by scholars in this field due to its high applicability and accuracy with a highly reliable quantization basis provided by big data. However, current data-driven methods typically provide only point estimates of the state rather than quantification of uncertainty, impeding effective maintenance decision-making. In addition, few studies have conducted further research on maintenance decision-making based on state predictions to achieve the full functionality of PdM. A PdM policy is proposed in this work to obtain the continuous probability distribution of system states dynamically and make maintenance decisions. The policy utilizes the Long Short-Term Memory (LSTM) network and Kernel Density Estimation with a Single Globally-optimized Bandwidth (KDE-SGB) method to dynamic predicting of the continuous probability distribution of the Remaining Useful Life (RUL). A comprehensive optimization target is introduced to establish the maintenance decision-making approach acquiring recommended maintenance time. Finally, the proposed policy is validated through a bearing case study, indicating that it allows for obtaining the continuous probability distribution of RUL centralized over a range of ±10 sampling cycles. In comparison to the other two policies, it could reduce the maintenance costs by 24.49~70.02%, raise the availability by 0.46~1.90%, heighten the reliability by 0.00~27.50%, and promote more stable performance with various maintenance cost and duration. The policy has offered a new approach without priori hypotheses for RUL prediction and its uncertainty quantification and provided a reference for constructing a complete PdM policy integrating RUL prediction with maintenance decision-making.

A Data-Based Tool Failure Prevention Approach in Progressive Die Stamping

The research on methods for monitoring sheet metal stamping is benefiting from the increased availability of enabling technologies such as sensors, data mining software, cloud computing, and artificial intelligence. The predictive maintenance policies of tools (punches and dies) can be targeted at monitoring progressive wear or at the detection of sudden failures or anomalies. Early detection of tool failure is the method preferred by the recent literature on data management in sheet metal stamping. However, the stamping of small parts poses challenges due to multiple tools and signals and limited visibility of die wear, requiring management of multiple sensors and data sources. This paper proposes a failure prevention approach for progressive die stamping using global and local force sensors with upper bounds for maximum values to indicate unhealthy conditions. The methodology was tested on millions of small washers made of carbon steel. The stamping process was implemented using a servo-press with a high rate. The press was equipped with eight in-process sensors, including strain gauges, thin foil force sensors, and acoustic sensors. The data of material properties, maintenance reports, statistical process control data, and in-process sensors were collected and stored in a data lake. By combining the in-process sensor acquisition with the corresponding log events and maintenance data in the same time span, it is possible to look for correlations among the variables and build an effective tool health prevention policy.

Integrating preventive and predictive maintenance policies with system dynamics: A decision table approach

A decision table is a practical tool that helps systems planners to make operational decisions, especially when they are under stress. With the effect of recent trends, such as the use of machine learning, data mining, and reinforcement learning methods, the maintenance decision has been a dynamic issue depending on system conditions. An expert may execute the maintenance or wait for the next periodic maintenance due to lack of maintenance workers, tools or budget, resources, etc., although the intelligent method predicts a failure approaching. Even sometimes, he/she may ignore the current periodic maintenance. Our method allows making some changes in the maintenance plan systematically. It integrates the results of preventive and predictive maintenance policies, and as different from the literature, it allows ignoring some maintenance actions depending on the maintenance resource levels in a decision table. Such a strategy helps to allocate limited resources to maintenance actions reasonably. We conducted an extensive simulation study on a real-life dataset. The preventive maintenance period is determined using classical approaches such as Weibull analysis. A machine learning algorithm is utilized to predict the type of failure. We have analyzed the performance of the proposed decision table approach under a variety of scenarios and with different parameter settings. We also showed the effect of parameter settings and the marginal utility of each maintenance policy. In addition, the approach provides several choices for planners. As a result, the proposed approach improves the system’s sustainability compared to traditional policies.

Simultaneous predictive maintenance and inventory policy in a continuously monitoring system using simulation optimization

Predictive maintenance (PdM) is a strategy which can help determine the appropriate timing for maintenance depending on the actual operational conditions of a manufacturing system. Many relevant studies have ignored the effects of resource dependency and only considered perfect maintenance to reduce the complexity and uncertainty of maintenance problems. In this study, we develop a novel and highly practical maintenance model to fill the gap. The proposed model simultaneously considers the predictive maintenance and inventory policies of spare parts. In light of the high degree of complexity in modern manufacturing systems and the profound stochasticity in component degradation, a metamodeling-based simulation optimization method is proposed to find the optimal component inventory policy and degradation level thresholds of each component. A numerical study is conducted to verify that the proposed model and solution method can find the optimal or nearly optimal maintenance cost effectively and efficiently. Furthermore, the impact of critical factors in the maintenance model on the optimal policy is analyzed and useful managerial insights are derived.

Machine learning (ML) based predictive maintenance policy for bridges crossing waterways

Scour is one of the major causes for bridge failures. This study established a machine learning based predictive maintenance policy for scour risk mitigation. Data used in this study was provided the French National Railway Company (SNCF). After comparing three commonly used ML algorithms, it is found that the extreme gradient boosting (XGBoost) algorithm outperforms others and achieves very encouraging results. Moreover, the trained classifier was then tested in another 40 bridge piers and the results were compared with a junior engineer's evaluation. Results show that XGBoost classifier had a slightly better performance in terms of accuracy and precision. The proposed methodology in this paper allows engineers to determine scour risk of bridge piers in an accurate yet rapid way.

The requirements of a value model for the strategic implementation of predictive maintenance

Predictive Maintenance (PdM) in industrial plants aims not only to reduce the failure rates of assets, but also to increment their useful life. However, additional costs must be faced when adopting such a policy, and granting that it will have a positive impact in the long term is not a trivial issue. The implementation of PdM should then be justified from a strategic perspective. Efforts have been made to develop models that provide guidance for the strategic implementation of PdM policies. However, these models generally consider only the economic feasibility of the policy, and not the contribution to value for the organization. Asset Management provides relevant principles for approaching this decision-making activity while following a value-based approach. This work aims to investigate the extent to which value-based decision-making has been implemented to drive the strategic implementation of PdM. The work concludes with the requirements that should be fulfilled by a value model to guide decision-making concerning PdM implementations.

Application of analytic hierarchy process (AHP) to evaluate maintenance practices in TVET institutions in Kenya

TVET institutions are expected to produce the bulk of Kenya’s middle level manpower equipped with skills to drive the attainment of infrastructural objectives of Kenya Vision 2030. However inadequate training resources brought about by poor management practices continue to impede efforts to equip Kenyan youth with employable technical skills. A programme to re-equip the institutions has been effected but it requires maintenance practices to be aligned with the new acquisitions to ensure they are constantly available for training. A survey was conducted through a questionnaire to engineering departments to evaluate compliance with maintenance performance indicators (PIs); a 5-point rating scale was used to score responses in the questionnaire that returned a combined Maintenance Management System Effectiveness Index of 0.517. This indicated a need for improvement of maintenance practices. The Analytical Hierarchical Process (AHP) method was applied which identified predictive maintenance policy as the best one to apply

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.

Weibull recurrent neural networks for failure prognosis using histogram data

Weibull time-to-event recurrent neural networks (WTTE-RNN) is a simple and versatile prognosis algorithm that works by optimising a Weibull survival function using a recurrent neural network. It offers the combined benefits of the sequential nature of the recurrent neural network, and the ability of the Weibull loss function to incorporate censored data. The goal of this paper is to present the first industrial use case of WTTE-RNN for prognosis. Prognosis of turbocharger conditions in a fleet of heavy-duty trucks is presented here, where the condition data used in the case study were recorded as a time series of sparsely sampled histograms. The experiments include comparison of the prediction models trained using data from the entire fleet of trucks vs data from clustered sub-fleets, where it is concluded that clustering is only beneficial as long as the training dataset is large enough for the model to not overfit. Moreover, the censored data from assets that did not fail are also shown to be incorporated while optimising the Weibull loss function and improve prediction performance. Overall, this paper concludes that WTTE-RNN-based failure predictions enable predictive maintenance policies, which are enhanced by identifying the sub-fleets of similar trucks.

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.

Simultaneous Predictive Maintenance and Inventory Policy in a Continuously Monitoring System Using Simulation Optimization

Simultaneous Predictive Maintenance and Inventory Policy in a Continuously Monitoring System Using Simulation Optimization

A novel decision support system for managing predictive maintenance strategies based on machine learning approaches

Nowadays, the industrial environment is characterised by growing competitiveness, short response times, cost reduction and reliability of production to meet customer needs. Thus, the new industrial paradigm of Industry 4.0 has gained interest worldwide, leading many manufacturers to a significant digital transformation. Digital technologies have enabled a novel approach to decision-making processes based on data-driven strategies, where knowledge extraction relies on the analysis of a large amount of data from sensor-equipped factories. In this context, Predictive Maintenance (PdM) based on Machine Learning (ML) is one of the most prominent data-driven analytical approaches for monitoring industrial systems aiming to maximise reliability and efficiency. In fact, PdM aims not only to reduce equipment failure rates but also to minimise operating costs by maximising equipment life. When considering industrial applications, industries deal with different issues and constraints relating to process digitalisation. The main purpose of this study is to develop a new decision support system based on decision trees (DTs) that guides the decision-making process of PdM implementation, considering context-aware information, quality and maturity of collected data, severity, occurrence and detectability of potential failures (identified through FMECA analysis) and direct and indirect maintenance costs. The decision trees allow the study of different scenarios to identify the conditions under which a PdM policy, based on the ML algorithm, is economically profitable compared to corrective maintenance, considered to be the current scenario. The results show that the proposed methodology is a simple and easy way to implement tool to support the decision process by assessing the different levels of occurrence and severity of failures. For each level, savings and the potential costs have been evaluated at leaf nodes of the trees aimed at defining the most suitable maintenance strategy implementation. Finally, the proposed DTs are applied to a real industrial case to illustrate their applicability and robustness.

A predictive maintenance policy considering the market price volatility for deteriorating systems

In reality, under the impacts of market dynamism and uncertainty, various input costs of maintenance optimization programs such as spare part cost strongly fluctuate over time. This paper aims at introducing the market price volatility (fluctuation) into the predictive maintenance optimization framework. To this end, we first adopt a Cox-Ingersoll-Ross (CIR) process to describe the evolution of interest rates and a log-diffusion process to model fluctuating maintenance costs. An adapted predictive maintenance policy is then proposed. According to the policy, maintenance decisions are made based on the predictive information about both the system health state and the price volatility level. The consideration of price volatility helps to reduce the negative impacts of the cost fluctuation on the maintenance decision-making, but it makes the evaluation of maintenance performance more complex. To overcome this issue, an analytical solution is developed based on the semi-regenerative properties of the maintained system. Finally, different theoretical and numerical studies confirm the necessity of taking into account the price volatility and the effectiveness of the proposed policy.

Energy Consumption-Based Maintenance Policy Optimization

The optimal predictive, preventive, corrective and opportunistic maintenance policies play an important role in the success of sustainable maintenance operations. This study discusses a new energy efficiency-related maintenance policy optimization method, which is based on failure data and status information from both the physical system and the digital twin-based discrete event simulation. The study presents the functional model, the mathematical model and the solution algorithm. The maintenance optimization method proposed in this paper is made up of four main phases: computation of energy consumption based on the levelized cost of energy, computation of GHG emission, computation of value determination equations and application of the Howard’s policy iteration techniques. The approach was tested with a scenario analysis, where different electricity generation sources were taken into consideration. The computational results validated the optimization method and show that optimized maintenance policies can lead to an average of 38% cost reduction regarding energy consumption related costs. Practical implications of the proposed model and method regard the possibility of finding optimal maintenance policies that can affect the energy consumption and emissions from the operation and maintenance of manufacturing systems.

Replacement and imperfect repair of deteriorating system: Study of a CBM policy and impact of repair efficiency

This paper studies a condition-based predictive maintenance policy in the presence of imperfect repairs for deteriorating systems. The effect of imperfect repairs is assumed to be random in the sense that a maintenance action reduces the systems degradation level by a random proportion. The associated distribution allows characterizing the repair efficiency. As an example, the beta distribution is considered. It allows defining a wide range of repair efficiencies. A maintenance policy with an aperiodic and predictive inspection planning is considered and a new decision criterion is introduced to evaluate the extent to which a repair would be beneficial. The aim is to find a proper balance between repair and replacement for preventive action which leads to optimal maintenance cost. With this consideration, the cost model is developed from regenerative and semi-regenerative properties. The expected cost is evaluated and the optimal policy is reached as a function of repair efficiency. Finally, a numerical study is drawn up to demonstrate the implementation and highlight the potential benefits of the proposed methods. A sensitivity analysis on critical parameters is also performed.

Decision support models for annual catenary maintenance task identification and assignment

A well-maintained catenary system is crucial to efficient and safe railway operations. Schedule catenary maintenance tasks with consideration of reliability and cost is vital for annual maintenance planning. Since past research on catenary maintenance planning mainly adopt the preventive maintenance (PM) policy with fixed maintenance intervals, this study considers practical concerns of railway operators and applies predictive maintenance (PdM) policy in the annual catenary maintenance planning. A decision support model is proposed by using both mixed integer programming and heuristic methods to identify and assign catenary maintenance tasks with the objective of minimizing maintenance cost and labor cost. The numerical results show that the cost can be improved by 25% compared to the current PM-only practice. The proposed model assists planners to determine and schedule maintenance tasks effectively and ensures the required reliability.

Data-driven dynamic predictive maintenance for a manufacturing system with quality deterioration and online sensors

Manufacturing systems are generally subject to product quality deterioration as a result of degradation of machine tooling-components, and thus tooling-component replacement has always been an effective way to ensure high product quality. For individual tooling components, online degradation data from sensors contain the information about their unique degradation patterns, which can be used to achieve precise degradation prediction. Therefore, this paper proposes a data-driven dynamic predictive maintenance policy which utilizes online degradation data to continuously enhance the degradation prediction of tooling components and further constantly revise the maintenance schedule. To account for individual heterogeneity, the tooling-component degradation rates are assumed as random variables, whose posteriori distributions are continuously updated using the online degradation data based on Bayesian approach. The real-time degradation prediction of tooling components is further used to predict the deterioration of product quality and machine reliability based on the response model and integrated hazard function. The quality and reliability deteriorations are measured by costs to construct a dynamic cost rate function, which is used to make adaptive PM schedules. Case study shows that the proposed policy can make practical maintenance schedules and achieve cost saving in general.