Time-based Maintenance Research Articles

Explainable, Deep Reinforcement Learning–Based Decision Making for Operations and Maintenance

This paper presents research that integrates condition monitoring and prognostics with decision making for nuclear power plant operations and maintenance aimed at reducing lifetime maintenance and repair costs. Additionally, a focal point of this research is to make the decisions explainable to operators, improving the trustworthiness of the decisions from what can be considered a black box model. In this work, we develop and evaluate an explainable, online asset management methodology to help reduce lifetime maintenance and repair costs. Using the latest advancements in condition monitoring, inventory management, deep reinforcement learning, and explainable artificial intelligence methods, we create a predictive maintenance methodology that can optimize the maintenance and spare part management of a repairable nuclear power plant system. To demonstrate these methods, preliminary studies were conducted on a representative maintenance system undergoing a stochastic degradation process that requires repairs or replacement to continue operation. Using deep reinforcement learning, we were able to reduce maintenance spending by approximately 50% compared to optimized, time-based maintenance strategies for the chosen system. A key component of our methodology is the integration of Shapley values to quantify the contribution of various factors to the decision-making process. This addition enhances the explainability and trustworthiness of our decisions, providing operators with transparent and understandable insights into the rationale behind maintenance strategies. The robustness and resiliency of our decision policy against observation noise were also thoroughly evaluated, demonstrating its effectiveness in uncertain operational environments.

Feasibility and foundational elements of CBM+ technology application in weapon systems

The maintenance strategies of weapon systems have evolved from corrective maintenance post-failure to time-based preventive maintenance, driven by the growing adoption of diagnostic technologies and data utilization. Currently, there is a transition to Condition-Based Maintenance (CBM) and, where applicable, Condition-Based Maintenance Plus (CBM+). This study investigated the feasibility of applying CBM+ technology using sensor data from weapon systems and explored methods for implementing the related foundational elements. First, the feasibility of CBM+ technology was confirmed by integrating data collection devices into the engine components of selected weapon systems, enabling sensor data collection and monitoring. Second, empirical evidence was gathered on the foundational elements necessary for the broader adoption of CBM+ technology. Practical application methods were proposed based on this evidence, including criteria for implementing CBM+ technology in major operational weapon systems, estimation of sensor data capacity, network methods for data transmission and collection, and the standardization of sensor data. The study’s findings are expected to provide reference points for selecting target weapon systems for the development and application of CBM+ technology, as well as outline essential foundational elements, such as security, networking, and data standardization, critical for leveraging related technologies.

Influence of the System Downtime Cost Rate on the Performance and Robustness of Periodic Inspection and Replacement and Quantitative Inspection and Replacement Maintenance Strategies Using Monte Carlo Method

Abstract This essay analyzes the crucial role of maintenance in production and service-oriented enterprises, highlighting the necessity for integrated viewpoints in contemporary corporate management. Focusing on time-based maintenance (TBM) and condition-based maintenance (CBM) techniques, the study provides a novel criteria and applies Monte Carlo simulation for assessment. The purpose is to discover ideal values for the system downtime cost rate, establishing a balance between performance and resilience. The organized parts include deterioration and failure models, maintenance assumptions, and cost models, finishing with a summary of noteworthy results and recommendations for further investigations. The study brings significant insights into the subject of maintenance strategies and analyzes the influence of the system downtime cost rate on the performance and robustness of condition-based maintenance strategies for a system with more or less stable behavior.

Phase-type distribution models for performance evaluation of condition-based maintenance

ABSTRACT Condition-based maintenance (CBM) is gaining attention due to sensor and cloud-based analytics advancements, but research on its impact on system-level performance is limited. Insufficient understanding during CBM implementation can lead to confidence issues and failures. This study introduces a class of models using phase-type distribution to assess three maintenance strategises: run-to-failure (RTF), time-based preventive maintenance (TBM), and CBM. Employing machine health-index, the framework characterizes production performance by estimating effective process times. The model demonstrates how adjusting CBM thresholds influences process time variations and assesses the impact of changing maintenance frequency for TBM. Applied to a smart cellular manufacturing system, the model shows CBM’s early-stage implementation. Findings indicate CBM with optimized thresholds boosts maximum throughput by 6.77%. Further, CBM achieves an additional 6.84% increase assuming corrective maintenance time can be reduced by 20%. This approach can help manufacturing become smarter through smarter maintenance in the Industry 4.0 era and beyond.

Developing data-driven O&M policy through sequential pattern mining: A case study

This study focuses on utilizing data mining techniques to extract valuable insights from discrete industrial data, crucial for Operations and Maintenance (O&M) decision-making. We present an innovative methodology integrating data mining with the Plan-Do-Check-Act (PDCA) cycle and Knowledge Discovery in Databases (KDD) principles. Through a manufacturing facility case study, we develop a hybrid opportunistic O&M policy, blending Condition-based Maintenance (CBM) and Time-based Maintenance (TBM) to replace the traditional Failure-based Maintenance (FBM) strategy. Our proposed maintenance policy incorporates delay-time modeling in a machining center’s critical sub-system, the lubrication equipment, proving its cost-effectiveness even amid parameter fluctuations. We assess the financial impact of adopting the Total Productive Maintenance (TPM) approach through Sensitivity Analysis, and compare it with established O&M policies, demonstrating the superior cost-effectiveness of our model.

Planning maintenance when resources are limited: a study of periodic opportunistic replacement

Abstract Accepted by: M. Zied Babai We study an aged-based replacement policy with two control limits. The first triggers opportunistic replacement and the second triggers a guaranteed replacement. The policy is novel because: the instances for component replacement are restricted to instances of time, which we call slots, that arise periodically; and a slot provides an opportunity for replacement with a particular probability. The policy models contexts in which maintenance is periodic, and resources are limited or execution of maintenance is not guaranteed. The policy is important for practice because it is simple and reflects the common reality of time-based maintenance planning. Long-run cost per unit time and average availability are calculated in a renewal-reward framework. Numerical study indicates that, if opportunities are rare, guaranteed replacement is beneficial and opportunities should be taken early in the life of a system. Using the policy, a maintainer can evaluate the cost–benefit of investing more resources to reduce the time between slots. Specific analysis and policy comparisons can be carried out using a web-application developed by the authors.

Implementation of Internet of Things for Condition-Based Maintenance in Elevator Systems: An Analytical Case Study

The aim of this paper is to show how IOT technology can allow highly distributed elevator equipment servicing by using remote-monitoring technology to ease a move from tradi- tional corrective maintenance (CM) and time-based maintenance (TBM) to more prophetic, condition-based maintenance (CBM) in evidence to accomplish different advantages.The Literature review can indicate that CBM has merit over conventional CM and TBM from a theoretical outlook, but it turns on constant monitoring enhancement via advanced IOT technology. A detailed case history was implement to bring practical proof that IOT allows elevator adamant to attain CBM.Through, a theoretical context, the CBM of elevators put together exper- tise.The venture lies in data collection, data analysis and decision making in real-world context of business. The key outcome of this research recommends that in the case of elevators the IOT can be commercialized and would improve the welfare and accuracy of equipment. It might, so, be understandable from technological, process and economic aspect. Our long term physical world case study reveal an actual way of building the CBM of elevators worldwide. Incorporating Internet of things and other modern technology will improve the safety and authenticity of elevator equipment,longer service life, minimize disruption or difficulty and business interference as a result of equipment downtime and to eliminate major repairs, thus result in low maintenance cost. A key contribution in this study stand in the empiric confirmation of the advantages and difficulties of CBM through Internet of things comparative to conventional CM and TBM in the instance of elevators. The writer had faith that this study is well timed and will be worth to company working on the same research or business strategies.These are form on condition based maintenance, IOT, Time-based maintenance and Corrective maintenance.

Developing and Implementing a Value-Based Operations and Maintenance Performance Management in Nigeria Electric Power Industry

The Nigeria electric power industry is yet to recognize and adjust management practices, especially those concerned with operations and maintenance, to the changed and changing business conditions in electric power production. Hence, there is a need to review the organization’s operations and maintenance policies. This paper intends to develop a value-based operations and maintenance performance which could be implemented in Nigeria electric power stations. An attempt was made to develop a framework for effective management of operations and maintenance performance which will link results to performance drivers. Furthermore, balance scorecard concept was included. The paper emphasized on value-added and intellectual capital in operations and maintenance in the emerging business environment, and stressed that there is a need to move from corrective maintenance or time-based maintenance to a long-term business-oriented approach.

Deep reinforcement learning for maintenance optimization of a scrap-based steel production line

This paper presents a Deep Reinforcement Learning (DRL)-based optimization approach for determining the optimal inspection and maintenance planning of a scrap-based steel production line. The DRL-based optimization maintenance recommends the adequate time for inspections and maintenance activities based on the monitoring conditions of the production line, such as machine productivity, buffer level, and production demand. Some practical aspects of the system, such as such uncertainty of the maintenance duration and the variable production rate of the machines, were considered. A scrap-based steel production line was modeled as a multi-component system considering components dependencies. A simulation model was developed to simulate the dynamics of the system and assist with the development of the DRL maintenance approach. The proposed DRL-based maintenance is compared with traditional maintenance policies, such reactive maintenance, time-based maintenance, and condition-based maintenance. In addition, different DRL algorithms such as PPO (Proximal Policy Optimization), TRPO (Trust Region Policy Optimization), and DQN (Deep Q-Network) are investigated in the case-based scenario. The findings indicated the potential for significant financial savings. Therefore, the proposed maintenance approach demonstrates system adaptability and has the potential to be a powerful tool for industrial competitiveness.

Advanced diagnostic techniques for turbo compressors: A spectral analysis approach for preventive maintenance

This paper introduces an innovative spectral analysis control approach aimed at monitoring and diagnosing machine malfunctions to prevent potential failures. The research was conducted on a critical machine in a major industrial enterprise. The proposed method involves the use of a new indicator, called Overall Level (OL), that evaluates the machine’s condition before any operation. This study showcases practical methodologies for transitioning from time-based maintenance to predictive strategies, furnishing actionable insights into machine condition. This yields tangible advantages for the industry in terms of optimizing maintenance practices and enhancing asset productivity. Additionally, various methods, including vibration analysis, performance monitoring, and data analysis, are employed to identify the causes of issues and recommend solutions to enhance the reliability of the turbo compressor. The results provide a clear representation of the machine’s vibration state for diagnostic purposes. This noteworthy intervention underscores the potential of incorporating the measured and calculated values of the OL indicator across three specifically chosen frequency bands. To achieve this objective, the average value is employed as an indicator, contributing to the enhancement of reliability and longevity of critical industrial machinery. In this context, the novelty of the findings resides in the advanced diagnostic capabilities of the turbocompressor, thereby augmenting the efficacy of condition-based preventive maintenance for the BP 103 J turbine. The ultimate goal is to extend the equipment’s lifespan, improve the efficiency of the rotating machinery fleet, reduce maintenance costs, and enhance parameters such as availability and reliability through the support of an electronic maintenance system.

An intelligent schedule maintenance method for hydrogen fuel cell vehicles based on deep reinforcement learning considering dynamic reliability

Schedule maintenance activities for hydrogen fuel cell vehicles are aimed to keep them in better condition in long-term operation. In this paper, we develop a reinforcement learning based schedule maintenance strategy, which is used to comprehensively consider the balance between safety and maintenance cost, so as to find the optimal maintenance strategy. A multi-level framework is established with remaining useful life of key components, fault tree analysis, and logistics cost model to generate an exploration environment that can express the operational stability, the failure rate of components, and the cost of repair and storage. The trained agent combines a deep neural network to explore the optimal strategy under the dynamic reliability of key components. Finite steps are the constraints. Safety rates, maintenance cost, and episode of operation are incorporated into a multi-objective reward function. The hydrogen supply circuit of fuel cell vehicle is simulated via Python. The trained agent is compared with traditional time-based schedule maintenance strategy and corrective maintenance strategy. The results show that the reinforcement learning based schedule maintenance agent has optimized the total reward, cost control and accident rate by 77%, 59% and 5%, respectively, compared with the traditional time-based schedule maintenance strategy. In the safe operation management of fuel cell vehicles, efficient and stable decision-making ability has been achieved.

УСПЕХИ В МОНИТОРИНГЕ СОСТОЯНИЯ КОНСТРУКЦИЙ: ОБЗОР ПОДХОДОВ МАШИННОГО ОБУЧЕНИЯ ДЛЯ ОБНАРУЖЕНИЯ И ОЦЕНКИ ПОВРЕЖДЕНИЙ

Structural Health Monitoring (SHM) is a crucial discipline geared towards detecting damage in engineering structures early, aiming to prevent failures and facilitate condition-based maintenance. Traditional SHM methodologies, relying on visual inspections, analytical models, and signal processing, exhibit inherent limitations. The advent of machine learning has introduced data-driven solutions to automate various aspects of SHM, including damage detection, localization, classification, and prognosis.
 This paper provides a comprehensive review of recent studies exploring supervised, unsupervised, and deep learning techniques in vibration-based, image-based, and multi-sensor SHM. Support vector machines, neural networks, deep convolutional neural networks, and other advanced algorithms have demonstrated exceptional performance in assessing damage using real-world structural datasets.
 Despite these successes, practical challenges persist, particularly in addressing variability and deploying machine learning models effectively on full-scale structures. Overcoming these challenges necessitates a more integrated, cross-disciplinary approach, merging mechanical engineering fundamentals with machine learning expertise. This synergy can pave the way for robust field implementation and further enhance the reliability of SHM systems.
 The transformative potential of machine learning in SHM cannot be understated. Beyond merely shifting from time-based maintenance to condition-based strategies, machine learning can automate and continuously evaluate structural integrity, ensuring the longevity of engineering structures. As we delve deeper into the intersection of mechanical engineering and machine learning, the prospect of a future where SHM seamlessly integrates with advanced technologies becomes increasingly tangible.

Residual life modeling and maintenance planning for repairable systems

Abstract Accepted by: Phil Scarf This paper formulates a state-dependent mean residual lifetime model for a repairable system operating in a dynamic environment. The problem is addressed by means of a two-state damage process reflecting the effect of operating environment on the system and a repair process associated with the damage process. As the damage process shifts to a higher state, to maintain a minimum level of performance, the decision maker repairs the system at times that arise according to a point process with a constant intensity. We demonstrate the generality of the proposed model and show how existing models emerge as specific cases. Our approach stimulates further research on the determination of two types of maintenance policies: maintenance policy based on the number of imperfect repairs (Model I) and time-based maintenance policy (Model II). In both cases, using the renewal reward theorem argument, we aim at minimizing the long-run average maintenance cost per unit time by determining optimal replacement policies and the optimal level of imperfect repairs. We illustrate the proposed models and carry out a comparative analysis of maintenance policies through numerical examples. The main conclusions drawn are that repair and maintenance policies depend on the failure mechanism, repair frequency and the level of costs involved. Also, numerical comparison shows that the maintenance modelling based on the number of imperfect repairs (model I) outperforms the time-based replacement model (model II) and two baseline models ignoring the effect of operating environment or whose attention is restricted to perfect repair.

A stochastic model of preventive maintenance strategies for wind turbine gearboxes considering the incomplete maintenance

In contemporary large wind farms, the combination of condition-based maintenance (CBM) and time-based maintenance (TBM) has become a prevalent approach in preventive maintenance, which is an indispensable part to ensure the safe, stable and environmental operation of equipment. However, the utilization of an inappropriate maintenance strategy may result in over-maintenance or under-maintenance, leading to unstable equipment operation. Furthermore, the majority of preventive maintenance involves replacement maintenance, which may have adverse effects on the performance of wind turbines with excessive maintenance time. Therefore, this paper takes the gearbox as a case study to introduce the incomplete maintenance parameters into the failure rate function to establish a state model based on the stochastic differential equation (SDE) and describing the state change of incomplete maintenance. And then simulating the state model of the gearbox and the joint preventive maintenance strategy of TBM and CBM through examples, resulting the time-based incomplete maintenance (TBIM) is proposed based on the TBM and the incomplete maintenance, and a new joint preventive maintenance strategy incorporating TBIM and CBM is proposed. Through developing the decision-making process of the maintenance strategy to optimize the inappropriate maintenance which including over-maintenance and under-maintenance and simulating the optimized preventive maintenance strategy to compare with that of TBM and CBM and verify the superiority and effectiveness of the proposed maintenance method.

Study on Vibration Adjustment Mechanism of S-76D Helicopter Rotor System Based on HUMS System

In recent years, helicopter maintenance technology has been developing rapidly and is transitioning from time-based periodic maintenance to condition-based maintenance. Helicopter Health and Usage Monitoring System (HUMS system) plays an important role in this process. The excitation forces originating from the helicopter rotor system are the primary cause of airframe vibration, and the HUMS system monitors the main rotor balance and trajectory data during each flight and provides an adjustment solution and predicts the vibration level in case of overruns. This paper summarizes the vibration measurement, cause analysis and vibration adjustment methods by studying the vibration adjustment mechanism of the S-76D helicopter rotor system, and provides solution for the same type of vibration problems of the helicopter.

Condition-based opportunistic maintenance strategy for multi-component wind turbines by using stochastic differential equations

The components of wind turbines are complex in structure and the working environment is harsh, which makes wind turbines face problems such as high failure rates and high maintenance costs. In this paper, the stochastic differential equation model has been established for the harsh operating environment of wind turbines, and used Brownian motion to simulate random disturbances; aiming at the problem of high failure rate of wind turbines, based on Weibull distribution, a new model has been established by combining operating time and equipment state to calculate the failure rate; in the analysis of monitoring data, the Higher-Order Moment method and Bayesian method were used to solve the parameters. The opportunity maintenance threshold curve and preventive maintenance threshold curve were obtained by analyzing Time-Based Maintenance and Condition-Based Maintenance. Therefore, the Condition-Based Opportunistic Maintenance strategy was obtained. The effectiveness of the proposed method was finally verified by arithmetic examples.

Comparative Analysis of Machine Learning Models for Predictive Maintenance of Ball Bearing Systems

In the era of Industry 4.0 and beyond, ball bearings remain an important part of industrial systems. The failure of ball bearings can lead to plant downtime, inefficient operations, and significant maintenance expenses. Although conventional preventive maintenance mechanisms like time-based maintenance, routine inspections, and manual data analysis provide a certain level of fault prevention, they are often reactive, time-consuming, and imprecise. On the other hand, machine learning algorithms can detect anomalies early, process vast amounts of data, continuously improve in almost real time, and, in turn, significantly enhance the efficiency of modern industrial systems. In this work, we compare different machine learning and deep learning techniques to optimise the predictive maintenance of ball bearing systems, which, in turn, will reduce the downtime and improve the efficiency of current and future industrial systems. For this purpose, we evaluate and compare classification algorithms like Logistic Regression and Support Vector Machine, as well as ensemble algorithms like Random Forest and Extreme Gradient Boost. We also explore and evaluate long short-term memory, which is a type of recurrent neural network. We assess and compare these models in terms of their accuracy, precision, recall, F1 scores, and computation requirement. Our comparison results indicate that Extreme Gradient Boost gives the best trade-off in terms of overall performance and computation time. For a dataset of 2155 vibration signals, Extreme Gradient Boost gives an accuracy of 96.61% while requiring a training time of only 0.76 s. Moreover, among the techniques that give an accuracy greater than 80%, Extreme Gradient Boost also gives the best accuracy-to-computation-time ratio.

Hydrogen in Glass Sector: A Comparison between Risk-Based Maintenance and Time-Based Maintenance Approaches

Hydrogen can be the key to decarbonisation, even for energy-intensive industries. The glass sector, for instance, burns natural gas to reach high temperatures necessary to melt the raw materials, leading to a considerable amount of CO2 emissions. Introducing hydrogen in a new sector brings many challenges in technological development, but of no less importance are the safety issues due to its hazardous properties. Hydrogen is highly flammable and can interact with many metals. Avoiding hydrogen leaks or, even worse, preventing catastrophic losses should be a priority: adopting proper maintenance planning is an effective means. In this study, a comparison of two different maintenance approaches is proposed in a line supplying hydrogen to the furnace case study. Time-based maintenance is a consolidated technique that programs the operations based on the reliability of the data on the pieces of equipment. On the other hand, the Risk-Based Maintenance approach leads to planning the maintenance activities based on the risk evaluation. Applying this approach to a hydrogen facility for the first time represents the novelty of this study. The advantages of adopting a methodology based on risk evaluation when handling a safety-critical system are highlighted.

AI-Driven Predictive Maintenance for Autonomous Vehicles for Product-Service System Development

The paper presents an Artificial Intelligence-driven approach to predictive maintenance for Product-Service System (PSS) development. This study focuses on time-based and condition-based maintenance, utilizing variational autoencoders to identify both predicted and unpredicted maintenance issues in autonomous haulers. By analyzing data patterns and forecasting future values, this approach enables proactive maintenance and informed decision-making in the early stages of PSS development.The inclusion of interaction terms enhances the model’s ability to capture the interdependencies among system components, addressing hidden failure modes. Comprehensive evaluations demonstrate the effectiveness and robustness of the developed models, showcasing resilience to noise and variations in operational data.The integration of predictive maintenance with PSS development offers a strategic advantage, providing insights into vehicle performance early in the development phases. This empowers decision-makers for efficient resource allocation and proactive maintenance planning. The research highlights the limitations and potential areas of improvement while also emphasizing the practical applicability and significance of the developed models in enhancing PSS development.

Designing a Resilient Building Maintenance Program: Integrating Preventive and Predictive Maintenance at the U.S. Navy

ABSTRACT Building equipment failure can have drastic effects on a company's operations and budget. This paper presents two types of maintenance approaches that if done effectively, can prevent or significantly reduce the failure of building equipment assets. The first is traditional time-based preventive maintenance (PM), which conducts prefailure inspections and tasks in a cyclic time-based approach. The second, is predictive maintenance (PdM), which conducts maintenance functions based on the condition of the equipment found through continuous or cyclic measurements and analysis during machine operation. The purpose of investigating these maintenance approaches is to determine whether we can improve on the U.S. Navy's (Navy) existing facility maintenance program, helping to reduce overall costs while improving sustainability, equipment resiliency, and efficiency. By presenting each maintenance program and leveraging today's technologies we show that these advancements in technology can directly improve the Navy's operational mission and warfighter readiness. Research was conducted through the following methods: interviews, books, third party reports, journal articles, industry websites and articles that focus on equipment maintenance. The Navy, the National Aeronautics and Space Administration (NASA), and the University of Washington provided case studies, existing facility/utility maintenance data, and budget information used in this research. The results show that PdM approaches using advanced analytics are more effective in diagnosing equipment, prescribing equipment problems, and predicting equipment failure. It will also show that when a PdM model is used, building tenants have less operational impacts as equipment operates longer with less downtime between maintenance events. By changing to a PdM program, facility managers and owners can improve asset efficiency and resilience, directly improving environmental sustainability and lowering overall longterm costs. It highlights the significant capital costs of a fully online PdM program and the benefits of using a hybrid model of PM and PdM. This research concludes with an overview of how building maintenance is currently being conducted on Navy bases by Naval Facilities Engineering Systems Command (NAVFAC), and how they are transitioning to a more sustainable maintenance program leveraging existing advanced metering infrastructure (AMI), building control systems (BCS), and utility control systems (UCS) with Smart Grid (SG), OSI Pi, and advanced analytics. In addition, major gaps in this transition are identified, and solutions are proposed to optimize these building system investments.