Optimal Maintenance Policy Research Articles

Maintenance strategy of ship multi-state deterioration system under reinforcement learning mode

Objectives Naval ship systems such as the hull structure, weapons equipment and power equipment will deteriorate during their service life. Thus, a ship maintenance strategy based on the actual deterioration state is essential for ensuring the safety and availability of naval ships. Methods In this paper, a multi-state deterioration system model is established on the basis of the Markov decision process. A reinforcement learning mode is then introduced to train the agent that generates the maintenance strategy, and the optimal condition-based maintenance strategy is obtained in the process of adaptive learning. Results The proposed method is applied to a ship structural deterioration system for demonstration, and the results show that it can obtain the optimal maintenance policy for a multi-state deterioration system considering the actual conditions, thereby providing an intelligent supporting tool for decision-makers to formulate optimal ship maintenance strategies. Conclusions This paper shows that the reinforcement learning method has great potential in comprehensively improving ship maintenance support.

Joint optimization of inventory planning, maintenance policy and pricing for perishable complementary products by considering the product freshness and technology level

Joint optimization of inventory planning, maintenance policy and pricing for perishable complementary products by considering the product freshness and technology level

Optimal inspection and maintenance planning for deteriorating structural components through dynamic Bayesian networks and Markov decision processes

Civil and maritime engineering systems, among others, from bridges to offshore platforms and wind turbines, must be efficiently managed, as they are exposed to deterioration mechanisms throughout their operational life, such as fatigue and/or corrosion. Identifying optimal inspection and maintenance policies demands the solution of a complex sequential decision-making problem under uncertainty, with the main objective of efficiently controlling the risk associated with structural failures. Addressing this complexity, risk-based inspection planning methodologies, supported often by dynamic Bayesian networks, evaluate a set of pre-defined heuristic decision rules to reasonably simplify the decision problem. However, the resulting policies may be compromised by the limited space considered in the definition of the decision rules. Avoiding this limitation, Partially Observable Markov Decision Processes (POMDPs) provide a principled mathematical methodology for stochastic optimal control under uncertain action outcomes and observations, in which the optimal actions are prescribed as a function of the entire, dynamically updated, state probability distribution. In this paper, we combine dynamic Bayesian networks with POMDPs in a joint framework for optimal inspection and maintenance planning, and we provide the relevant formulation for developing both infinite and finite horizon POMDPs in a structural reliability context. The proposed methodology is implemented and tested for the case of a structural component subject to fatigue deterioration, demonstrating the capability of state-of-the-art point-based POMDP solvers of solving the underlying planning stochastic optimization problem. Within the numerical experiments, POMDP and heuristic-based policies are thoroughly compared, and results showcase that POMDPs achieve substantially lower costs as compared to their counterparts, even for traditional problem settings.

Structure of an optimal maintenance policy for a semi‐Markovian deteriorating system with major and minor failures

AbstractWe consider a multistate system wherein deterioration is modeled as a semi‐Markov process with an absorbing state. The system can suffer major and minor failures. When the system reaches the absorbing state, major failure occurs and it is replaced. Meanwhile, minor failure can occur at each state. Upon occurrence of a minor failure, the system is minimally repaired. We propose an optimal state–age‐dependent maintenance policy for the system which minimizes an expected long‐run cost rate. We study structural properties of the optimal maintenance policy and show that a control‐limit policy holds under several assumptions. We also derive a special relation focused on minor failure related to the state–age‐dependent maintenance policy.

Optimal Maintenance Policy for Offshore Wind Systems

Employing maintenance threshold plays a critical step in determining an optimal maintenance policy for an offshore wind system to reduce maintenance costs while increasing system reliability. Considering the limited works on this topic, we propose a two-stage procedure to determine the optimal maintenance thresholds for multiple components of an offshore wind power system in order to minimize maintenance costs while achieving the highest possible system reliability. First, using genetic algorithms, a dynamic strategy is developed to determine the maintenance thresholds of individual components where the cost of maintenance and the rate of failure are critical. Then, fuzzy multi-objective programming is applied to find the system’s optimal maintenance threshold considering all components. A variety of factors including weather conditions, system reliability, power generation losses, and electricity market price are carefully considered to enhance the system’s reliability and reduce the costs of maintenance. When maintenance threshold results are compared, component-wise versus system-wise, an average system savings of 1.19% for maintenance cost is obtained while the system reliability is increased by 1.62% on average.

A risk-aware maintenance model based on a constrained Markov decision process

The Markov Decision Process (MDP) model has been widely studied and used in sequential decision-making problems. In particular, it has been proved to be effective in maintenance policy optimization problems where the system state is assumed to continuously evolve under sequential maintenance policies. In traditional MDP models for maintenance, the long-run expected total discounted cost is taken as the objective function. The maintenance manager’s target is to evaluate an optimal policy that incurs the minimum expected total discounted cost through the corresponding MDP model. However, a significant drawback of these existing MDP-based maintenance strategies is that they fail to incorporate and characterize the safety issues of the system during the maintenance process. Therefore, in some applications that are sensitive to functional risks, such strategies fail to accommodate the requirement of risk awareness. In this study, we apply the concept of risk-aversion in the MDP maintenance model to develop risk-aware maintenance policies. Specifically, we use risk functions to measure some indexes of the system that reflect the safety level and formulate a safety constraint. Then, we summarize the problem as a constrained MDP model and use the linear programming approach to evaluate the proposed risk-aware optimal maintenance policy under concern.

Maintenance optimization for a single wind turbine component under time-varying costs

In this paper, we introduce a new, single-component model for maintenance optimization under time-varying costs, specifically oriented at offshore wind turbine maintenance. We extend the standard age replacement policy (ARP), block replacement policy (BRP) and modified block replacement policy (MBRP) to address time-varying costs. We prove that an optimal maintenance policy under time-varying costs is a time-dependent ARP policy. Via a discretization of time, the optimal time-dependent ARP can be found using a linear programming formulation. We also present mixed integer linear programming models for parameter optimization of BRP and MBRP. We present a business case and apply our policies for maintenance planning of a wind turbine gearbox and show that we can achieve savings up-to 23%.

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.

Determining the Optimal Maintenance Strategy for Ammonium Hydroxide ‎Production Unit Using Risk-Based Inspection ‎and Analytic Hierarchy Process

Ammonium Hydroxide production ‎unit (AHPU) is one of the widely ‎used and important units available in ‎many ‎industries such as oil and gas, ‎petrochemical, and power plants. In ‎this research, for the first time, a ‎detailed systematic inspection and ‎maintenance plan for AHPU ‎is ‎determined based on the ‎prioritization of equipment using ‎equipment’s risk analysis. ‎Equipment’s failure risk is ‎calculated ‎based on probability and ‎consequence of failure according to ‎American Petroleum Institute ‎Recommended Practice (API RP) 581. ‎Different ‎maintenance strategies were ‎designed and compared considering ‎several criteria like safety, cost, and ‎feasibility using the analytic hierarchy ‎process. Then risk-based optimal ‎maintenance policy for each group of ‎equipment ‎was identified. Finally ‎results showed that, the equipment ‎identified as high or medium to high ‎risk level such as liquid ‎ammonia ‎storage tank and its related ‎pipes, should has preventive ‎maintenance with an inspection ‎period of 48 months. While for the ‎medium risk equipment such as ‎ammonium hydroxide storage ‎tank ‎and low-risk equipment such as ‎ammonia absorption tank, should ‎have corrective maintenance with ‎inspection period of 72 and 90 ‎months, respectively‎.

Data-driven modelling, analysis and improvement of multistage production systems with predictive maintenance and product quality

Predictive maintenance (PM) and quality management help to improve the business bottom line by alleviating the system performance degradation caused by unscheduled machine breakdown and product quality problems. In modern production systems, the wide application of new IT technology results in data-rich environments. However, it is not clear how to take advantage of the data to facilitate maintenance decision-making and production performance improvement. Aiming at multistage production systems with batching machines and finite buffers, this research studies data-driven modelling, analysis and improvement of production systems with predictive maintenance and product quality. First, a data-driven quantitative method is proposed to analyze the impact of machine breakdowns, predictive maintenance and product quality failure on system performance. Then, based on the obtained system production loss, a PM decision model is established to minimise the maintenance and production costs, and the optimal maintenance policy is exploited based on an approximate dynamic programming algorithm. In addition, downtime bottleneck (DT-BN) is defined, and a data-driven bottleneck indicator is derived. A continuous improvement method is established through the identification and mitigation of the bottlenecks. Finally, numerical case studies are performed to validate the effectiveness of the proposed PM decision model and continuous improvement method.

Optimal Age-Based Preventive Maintenance Policy for a System Subject to Cumulative Damage Degradation and Random Shocks

This research investigates the problem of cumulative degradation and random shocks a system like a centrifugal pump may experience during normal and adverse operating conditions. An accelerated life testing method was employed to determine the degradation of the pump under cumulative damage degradation and random shocks conditions. An age- Based policy was used to determine the optimum time interval that will minimize the total expected cost of the system. The random shock increases the number of failures and hence reduces the reliability of the system. The total expected preventive maintenance cost obtained varies from N1700.00 (One thousand seven hundred naira) to N16,000.00 (sixteen thousand naira), depending on the shock and shock duration. The methodology presented is useful and thus recommended for use to study cumulative damage degradation and random shocks for similar systems.

Optimal Maintenance Probabilities and Preventive Replacement Maintenance Policy for Photocopy Machines

The need for suitable replacement policies are essential to minimize down time, maintenance cost and maximize the availability and reliability of equipment. On this premise, this work models the failure rate of Photocopy machines and obtain its optimal preventive maintenance policy that would prevent damage and its attendant losses to both users and end-product consumers. The failure distribution of the machine was shown to follow the Log-Logistic distribution with shape parameter, ˆα=1.723339368 and scale parameter, ˆβ=763.9219635. Optimal probabilities of the distribution were obtained and utilized in both the cumulative failure function and cumulative hazard function-based replacement models to formulate a replacement maintenance policy for the machine. The failure cumulative function-based replacement model was found to be a better model which yields optimal replacement maintenance time of 166 hours at a minimum cost of 113 Naira for maintaining the machine per cycle time with 96% availability, 94% reliability and 0.07% chance of failure occurrence in the machine.

Optimal control-limit maintenance policy for a production system with multiple process states

In this paper, a new control limit maintenance policy is proposed for a production system with multiple process states. The process state can be obtained by quality information or other measurable indicators from the production process, when the process state by itself provides an insufficient indication or measure of the actual machine state. This can make the maintenance activities difficult to implement. Alternatively, machine virtual age can indicate the actual machine state and the maintenance effect on the machine state. Therefore, machine virtual age is used to compensate for the possible weakness of process state as an indication of machine state, and it is also used as part of maintenance decisions for more effective maintenance activities. At each inspection time, maintenance decisions are made for spare parts ordering and maintenance activities. The maintenance activities which are considered in this paper are replacement and two types of imperfect maintenance. The new maintenance planning model considers the different maintenance levels and their effects on system performance. Moreover, different maintenance cost and durations are considered. The system and associated maintenance policy are modeled by a discrete-time Markov decision process, and the long-term expected cost rate is investigated. A modified policy-iterative algorithm is then proposed to accelerate the search speed effectively during cost optimization. Compared to a traditional algorithm, the modified algorithm checks the effects of some possible parameters, rather than all parameters during the policy-iteration process. Simultaneously, it assures that the checked parameters are better than those unchecked. Finally, a numerical example is presented to illustrate the proposed method, and sensitivity analyses of important parameters are also implemented.

An optimal opportunistic maintenance policy for a two-unit series system with general repair using proportional hazards models

This paper proposes an opportunistic maintenance policy for maintaining mechanical systems consisting of two series units with economic dependency. Condition monitoring and age information is available for unit 1 and unit 2, respectively. A gamma process describes the stochastic deterioration of unit 1, and condition monitoring data is utilized to estimate its hazard rate using the proportional hazards model (PHM). A matrix-based algorithm is employed to derive the conditional reliability of a two-unit series system with a continuous-state deterioration process. The policy proposes preventive actions on scheduled intervals, based on lifetime and PHM parameters, to enhance system reliability and mitigate unexpected downtimes. Corrective actions rectify the failed unit, while the other unit can be maintained opportunistically. The problem is formulated as a semi-Markov decision process (SMDP) with an objective of minimizing the long-run expected average cost over an infinite time horizon. This paper relaxes a common assumption that repair brings the units to the as-new condition, and provides solutions for the maintenance scheduling of two-unit series systems, where repair is also allowed.

Reliability modelling based on dependent two-stage virtual age processes

This paper proposes reliability and maintenance models for systems suffering random shocks arriving according to a non-homogeneous Poisson process. The system degradation process include two stages: from the installation of a new system to an initial point of a defect (normal stage), and then from that point to failure (defective stage), following the delay time concept. By employing the virtual age method, the impact of external shocks on the system degradation process is characterized by random virtual age increment in the two stages, resulting in the corresponding two-stage virtual age process. When operating in the defective state, the system becomes more susceptible to fatigue and suffers from a greater aging rate. Replacement is carried out either on failure or on the detection of a defective state at periodic or opportunistic inspections. This paper evaluates system reliability performance and investigates the optimal opportunistic maintenance policy. A case study on a cooling system is given to verify the obtained results.

Reliability analysis of periodically inspected systems with competing risks under Markovian environments

We develop a reliability and maintenance model for a periodically inspected system with competing risks subject to a randomly evolving operating environment, which is modeled as a continuous-time homogeneous absorbing Markov process. The natural wear rate and shock arrival rate are both modulated by Markovian environments. Specifically, three competing failure processes are considered. System failures may occur when an extremely harsh environment arrives, when the magnitude of a shock load exceeds a fixed value, or when the overall cumulative degradation caused jointly by the natural wear and abrupt additions due to shocks hits a preset threshold. Explicit formulas for reliability functions and first failure time moments of systems are derived. A Monte-Carlo simulation algorithm to obtain the first failure time of the system when shock arrivals are governed by non-homogenous Poisson processes is proposed. An optimal maintenance policy using periodic inspection is developed by minimizing the average long-run cost rate where the decision variable is the inspection interval. A case study of micro-engine systems is given to illustrate the developed reliability and maintenance model, which exhibits that an increased failure threshold value improves the system reliability and reduces the minimum average long-run cost rate.

Joint optimization of maintenance and inventory policies for multi-unit systems

A maintenance policy and the management of the inventory of spare parts and their joint optimization often challenge managers and researchers. In this paper, the first analytical joint optimization model is established. A simulation model is then developed for the system operating under the suggested condition-based maintenance to optimize the maintenance outline of mining dump truck motors based on oil monitoring. Our model is combined with a genetic algorithm to obtain the optimal response. In the presented model, the Inspection intervals ( $$T$$ T ) and the maximum stock level ( $$S$$ S ) are jointly optimized for minimizing cost. To build a sample and a simulation of various repair events, 11,000 oil analysis data is used. The deterioration of spare parts is shown with an increasing numerical variable over time, which follows a function. Using the existing datasets, the deterioration rate function is obtained. Another process required for simulation is the failure probability function. Due to the extent of deterioration with various breakdowns, there are uncertainties and different values. Condition-based maintenance is used to determine the deterioration level of failure. In the end, the results of the simulation are compared with the current costs resulting from the workshop policies.

Availability analysis and maintenance optimization for multiple failure mode systems considering imperfect repair

This paper studies the availability and optimal maintenance policies for systems subject to competing failure modes under continuous and periodic inspections. The repair time distribution and maintenance cost are both dependent on the failure modes. We investigate the instantaneous availability and the steady state availability of the system maintained through several imperfect repairs before a replacement is allowed. Analytical expressions for system availability under continuous and periodic inspections are derived respectively. The availability models are then utilized to obtain the optimal inspection and imperfect maintenance policy that minimizes the average long-run cost rate. A numerical example for Remote Power Feeding System is presented to demonstrate the application of the developed approach.

Joint optimization of preventive maintenance and inventory management for standby systems with hybrid-deteriorating spare parts

As an effective way to enhance the system reliability, the standby redundancy technique has been widely applied in many industrial systems. Finding methods to determine the maintenance policy and spare-part inventory management is an interesting practical issue. However, due to the imperfect storage and inner mechanism, the spare parts usually deteriorate over time, which not only degrades their performance but also may lead to storage failure. In addition, the immediate burst failure caused by the external shock should not be neglected in storage. These two aspects make the joint optimization of preventive maintenance and inventory policies for a standby system more challenging. Thus, in this study, we first propose a general iterative approach for the lifetime estimation of standby systems with a hybrid spare-part degradation process, which includes both stochastic degradation and immediate burst failure. We then take an example of the Wiener-process-based model to explain how to obtain the analytical results of the lifetime estimation. On this basis, we further establish a joint optimization model, in which the preventive block replacement interval and the inventory number are treated as two decision variables to minimize the expected cost per unit time. Finally, a numerical example and practical case are provided for illustration.

Incorporating Climate Change in Pavement Maintenance Policies: Application to Temperature Rise in the Isfahan County, Iran

Although temperature rise is imminent in Iran and could damage asphalt pavements, no national guide exists to adapt them. To ensure the sustainability of pavements against temperature rise, a county-level methodology based on machine-learning algorithms was developed. To show the applicability of the framework, the Isfahan County was studied. The county's climate was found to change from cold-semi-desert to relatively-warm-semi-desert in future decades. Then, optimal maintenance policies before and after climate change were identified. It was concluded that optimal policies of arterial roads before and after climate change were more intense than those of local roads. Furthermore, optimal policies after climate change were more intense than those before climate change at additional costs of 1379.57 MR/KM and 632.49 MR/KM respectively for arterial and local roads. The same methodology could be applied to sustainably adapt asphalt pavements of other counties. To validate the research, a questionnaire survey of pavement management and climate change experts was done. The experts confirmed that the methodology facilitates achieving sustainable development goals #9, #11, and #13 by improving maintenance budget allocation, enhancing policy-makers communication with authorities, maintaining adequate technical and end-user levels of service, and adapting pavements to climate change through cost-effective and performance-effective maintenance policies.