Search

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.

International audience; In recent decades, the technology and techniques for condition monitoring have experienced a rapid development. However, there is still a need for reducing unnecessary inspections and/or preventive maintenance actions and their associated cost, through optimal design of condition-based maintenance (CBM) strategies. Accordingly, mathematical modelling and optimization of CBM has become of interest for industry managers and researchers. This work explores on the application of a piecewise deterministic Markov process (PDMP) to encompass different modelling assumptions as non-negligible maintenance delays and inspection-based condition monitoring. These assumptions are relevant for many critical infrastructures in civil engineering or in oil & gas industry whose deterioration states are classified at a very high level of abstraction among a finite and small set of possible states. A formalism to model this type of problems is proposed in which the deterministic motion of the PDMP is reduced to a trivial differential equation to track the time elapsed between events. A numerical scheme for quantification, as an approximation of the Chapman–Kolmogorov equation, is presented. Later, an illustration case dealing with CBM of road bridges by the NPRA (Norwegian Public Roads Administration) is presented, guiding through the modelling and quantification approach.

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.

International audience; The quality of condition monitoring is an important factor affecting the effectiveness of a condition-based maintenance program. It depends closely on implemented inspection and instrument technologies, and eventually on investment costs, i.e., a more accurate condition monitoring information requires a more sophisticated inspection, hence a higher cost. While numerous works in the literature have considered problems related to condition monitoring quality, (e.g., imperfect inspection models, detection and localization techniques, etc.) few of them focus on adjusting condition monitoring quality for condition-based maintenance optimization. In this paper, we investigate how such an adjustment can help to reduce the total cost of a condition-based maintenance program. The condition monitoring quality is characterized by the observation noises on the system degradation level returned by an inspection. A dynamic condition-based maintenance and inspection policy adapted to such a observation information is proposed and formulated based on Partially Observable Markov Decision Processes. The use and advantages of the proposed joint inspection and maintenance model are numerically discussed and compared to several inspection-maintenance policies through numerical examples.

International audience; Preventive maintenance planning is an important problem for the handling of energy production systems with high down time costs. Throughout the last decade different maintenance strategies have been developed and optimized in order to minimize operational and maintenance costs whilst conserving and improving the system reliability and productivity. Preventive maintenance strategies are usually based on the monitoring and the prediction of the system behavior and its deterioration process. However, some industrial systems may be operating under a dynamic environment and/or variable working conditions. In this case both the deterioration and the production processes may not be deterministic and incorporate different types of uncertainties. In this paper, we consider the case of a preventive maintenance strategy for a production system subject to uncertainty. For this system, a decision-making procedure for condition-based maintenance planning is proposed. In order to consider uncertainty in production and deterioration processes, these latter are modeled by non-monotonic stochastic processes. The modeling of deterioration processes by means of jump-diffusion stochastic processes has been proposed in our previous work. In this paper, a decision-making approach for preventive maintenance strategies is proposed. Knowing the remaining useful life of a system, a simulation-based real options analysis is used in order to determine the best date to maintain. Considering a case study of a wind turbine with PHM structure, the decision-making approach is described and tested through an empirical example.

International audience; Health prognosis is an advanced process to forecast the future state of systems, structures and components. Even if it is now recognized as a key enabling step for the maintenance performance improvement on systems and structures, the issue of post-prognosis maintenance decision-making (i.e., how to use prognosis results to eventually make maintenance decisions) remains open. Faced with this situation, we propose a parametric predictive maintenance decision framework that can take into account properly the system remnant life in maintenance decisions. Unlike more classical frameworks, it uses the estimated precision on the prognosis of the system residual useful life as a condition index to decide for and to schedule the interventions on the system. The proposed framework is developed for a single-unit stochastically deteriorating system, maintained through inspection and replacement operations. Using results from the theory of semi-regenerative phenomena, the analytical maintenance cost model is derived for the long-run expected maintenance cost rate. The proposed maintenance decision structure is compared to a classical benchmark framework; numerical experiments evidence the performance and the robustness of the new framework, and confirm the benefit of basing maintenance decisions explicitly on the precision of the system health prognosis (and not only on e.g., the mean value of the estimated residual life).

International audience; This paper is devoted to proposing a flexible continuous wind speed model based on mixtures of Markov chain and stochastic differential equations. The motivations are as follows: (a) the model is intended to generate wind speed sequence with statistical properties similar to the observed wind speed for a particular site. (b) The model is intended to generate long-term wind speed patterns with high frequency when necessary. This model is flexible enough to generate wind speed at different timescales without time-length limitation. Different stochastic differential equations which express various environmental conditions and behaviors can be used in the model, make the proposed model adapt for various configurations. Two forms of stochastic differential equations have already been studied in this paper; other forms could be applied to this model according to the user's requirement. A sequence of generated wind speed is applied to a wind turbine simulator as an application example. Having low computational requirement is another advantage of the model.

International audience; For the very first time a jump diffusion process with Markovian covariates is used for prognostics and health monitoring. This paper aims to give a methodology for reliability and lifetime estimation of a non-increasing, deteriorating system where the increments are not normally distributed and the Wiener process cannot be used. We propose, first to study of the statistical properties of increments. Afterward the measures dependency is tested. A jump diffusion process with a covariate is considered to fit the deterioration model. The covariate models the environmental conditions. The proposed deterioration model is applied to a large set of collected data corresponding to the health indicators in ageing components in hydraulic electrical production systems and prognostic results are given.

International audience; As a consequence of high failure rate occurring to wind turbines (WTs), more and more asset owners begin to pay attention to WT’s maintenance planning. To implement an efficient condition-based maintenance policy, the deterioration modeling is of primary importance. The inherence character of wind makes WT’s operation full of uncertainties, hence, the pitch actuator movement is random and it can be modeled by a stochastic process. Meanwhile, the deterioration of the pitch control system related to the usage of the actuator is random. When the wind speed is higher than the rated wind speed (a design WT parameter, when wind speed is higher than the rated wind speed, the pitch system begin to operate) and the wind speed turbulence intensity (TI) is very important, the deterioration rate of the pitch control system increases significantly. For instance the pitch system has more failures in a windy season. Therefore, in order to propose an appropriate maintenance policy for pitch control system, it is more sensible to take into account the influence of wind speed in deterioration modelling.In this paper, we focus on the deterioration of hydraulic pitch control system. After studying the factors influencing the deterioration, a stochastic process considering wind speed as covariates is used to model the deterioration. A maintenance policy based on an alarm threshold is considered.

International audience; Health monitoring data are vital for failure prognostic and maintenance planning. Continuous monitoring data or frequent inspections can provide a large amount of information on degradation evolution and therefore ensure the quality of deterioration modeling and the lifetime prognostic accuracy. However, they are usually very costly, and sometimes inpractible in real engineering applications. Therefore, it is essential to address the issue of the appropriate amount of monitoring data. This paper proposes a new methodology to help the companies improving their actual inspection/monitoring policy to reduce operation and maintenance costs but also ensure the information quality. We investigate different types of inspection policies including periodic or non-periodic ones by considering multiples functions of the system degradation state that are linear, concave or convex. The best policies are chosen based on a multiobjective optimization problem dealing with the inspection cost and the information level. The advantages and disadvantages of the proposed methodology are discussed through numerous numerical examples for different types of degradation process, particularly Wiener and Gamma processes that have been largely addressed in the framework of degradation modeling.

This paper deals with the problem of the reliability certification for an autonomous driving (AD) system and provides some insights for assessing the system reliability in a minimum testing period. Preliminary works [1] propose a decomposition of the system reliability as a function of known and unknown driving scenarios. Here, the method focuses on the reliability assessment when the driving situations are assumed to be identified and partially known. A Bayesian framework is designed for mixing numerical simulation based and field data obtained during open road tests. Moreover, for reducing the number of kilometers in the open road tests, a selection of the “most critical” scenarios should be done. A simplified real-world example has been designed to illustrate the efficiency of the methodology in terms of speed of convergence for the validation process.

ACL; International audience; This paper analyzes three time-based replacement policies when the parameters of the time-to-failure distribution are unknown. Under the hypothesis of a large sample data, the unknown parameters are estimated via maximum likelihood method. The uncertainty around the point estimates impacts the maintenance decision rules. Based on the asymptotic normality of the parameter estimators and delta-method, the asymptotic distribution of the optimal decision parameter and the average cost is derived. This distribution permits to measure the uncertainty around the optimal maintenance decision rule induced by the point estimates. The sensitivity analysis of the replacement policies to the maintenance unit costs and lifetime distribution parameters is carried out. Application to two time-to-failure distributions is provided.

International audience; Degradation modeling and RUL estimation of systems are two essential problems in the research field of prognostics and health management. In the literature, one usually deals with these problems in the time domain. Here, we propose another perspective on this subject by studying the degradation modeling and RUL estimation in the s-plane. The study is based on a well-known mass-spring-damper deteriorating system. A complete procedure consisting of data collection, degradation indicator implementation, parameters estimation, failure definition and RUL estimation for the considered system is outlined. Some numerical experiments have been carried out to illustrate the procedure.

International audience; Over the last few decades, many efforts have been invested in improving the economic performances of maintenance policies for stochastically deteriorating production systems. However, with the development of complex production systems, maintenance managers are interested not only in cost saving, but also in how to trustworthily plan and allocate the required maintenance budget. In this context, the robustness of maintenance policies which is related to the maintenance cost variability from a renewal cycle to another plays a pivotal role. This research deals with a quantitative approach to jointly assess the economic performance and robustness of some representatives of two most well-known classes of maintenance policies: time-based and condition-based maintenance. To this end, we first propose a new cost criterion which combines the long-run expected cost rate and standard deviation of maintenance cost per renewal cycle. Then, we develop and compare the associated mathematical cost models of the considered maintenance policies on the basis of the Gamma degradation process and the theory of stochastic renewal processes. The comparison results under different situations of maintenance costs and system characteristics show that the optimal configuration of maintenance policies gives the best compromise between the performance and robustness, and is mostly affected by the system downtime. Under this aspect, the condition-based maintenance remains more profitable than the time-based maintenance. Still, maintenance managers could implement condition-based maintenance policies that efficiently control the downtime to maximise the maintenance effectiveness of production systems from both performance and robustness viewpoints.

International audience; Seeking condition indices characterizing the health state of a system is a key problem in condition-based maintenance. For this purpose, diagnostic and prognostic models have been unceasingly developed and improved over the past few decades; nevertheless none of them explains thoroughly the impacts of such indices on the effectiveness of maintenance operations. As a complement to these efforts, this paper analyzes the effectiveness of some well-known diagnostic and prognostic indices for maintenance decision-making. The study is based on a system subject to competing risks due to multiple crack paths. A periodic inspection scheme is used to monitor the system health state. Each inspection returns the perfect diagnostic information: the number of cracks, corresponding crack sizes, and the system failure/working state. Based on this information, two kinds of prognostic condition indices are predicted: the average value and probability law of the system residual useful life. The associated condition-based maintenance strategies and cost models are then developed and compared with the ones whose maintenance decisions are based on diagnostic condition indices. The comparison results allow us to conclude on the performance and on the robustness of these strategies, hence giving some suggestions on the choice of reliable condition indices for maintenance decision-making.

This paper tackles the maintenance decision-making of gradually deteriorating control systems. The main feature of these systems is their ability within a closed-loop to drive actuators in order to set the controlled process in a given state. Usually, the existing literature deals with the diagnosis of faults of a control system and the means to recover system performances after their appearance. The controller reconfiguration is one of these means. The root cause of a fault is rarely argued nor its occurrence time. Before designing maintenance policies, this paper proposes a stochastic modeling framework of a degrading control system focusing on the actuator deterioration. It is based on the assumption of a close relationship between the controller setting, the actuator deterioration and finally its faulty situations. These latter are related to given degradation thresholds. Due to the stochastic nature of the deterioration, the thresholds hitting times are random. The obtained model allows the conditional reliability of the actuator to be assessed and then the prognosis of its residual useful life (RUL). This RUL is finally used to achieve two maintenance policies based upon the reconfiguration of the controller which is therefore considered as a new maintenance action. © 2017. This is the authors' manuscript to the article. The final authenticated version is available online at: http://journals.sagepub.com/doi/10.1177/1748006X17692155

International audience; Maintenance and inventory optimizations are two interrelated processes but often investigated separately in the literature. Joint optimization of these processes can reduce the total maintenance and inventory cost, it may however lead to a complex problem with a large numbers of decision parameters to be optimized. To face this open issue, in this paper, we present a joint predictive maintenance and inventory strategy for systems with complex structure and multiple non-identical components. Both predictive maintenance and spare parts provisioning operations are studied and optimized jointly. The prognostic condition index and the structural importance measure of components are jointly used to build the thresholds for preventive maintenance and spare part ordering. Additionally, in order to take the advantages of economic dependence, opportunistic maintenance decision rules based on the criticality level of components and their spare parts availability are proposed. To evaluate the proposed joint strategy, a cost model taking into account the economic dependence of both maintenance and inventory activities is developed. A numerical study of a 7-component system is finally presented to illustrate the use and the advantages of the proposed strategy. Indeed, this strategy shows the flexibility and the efficiency in joint maintenance and inventory optimization.

International audience

International audience; The paper treats the problem of the maintenance of controlled systems. The actuator of the system is subject to deterioration due to wear and to shocks caused by environmental changes. This latter is modeled by random variables called covariates. The deterioration of the actuator impact the control system and requires more effort to balance the deterioration impact and the controlled out put of the system. The main purpose of this paper is to propose a maintenance policy based on the residual lifetime of the system regulated by a Markovien command by taking into account the deterioration of the actuator. The originality of the paper is in one hand due to the consideration of a random deterioration of the actuator in the framework of a controlled system, and on the other hand the consideration of the remaining useful lifetime in the maintenance decision rule.

International audience; In this paper, how to estimate residual useful lifetime (RUL) is discussed based on a model-based method when fluctuations exist in the degradation process around its average behavior. The main idea is to model the degradation process as a time-dependent Ornstein-Uhlenbeck (OU) process, where the first passage failure is adopted to consider corresponding RUL estimation. The time-dependent OU process is proved good by its statistical properties on the controllable mean, variance, and correlation. Its mean-reverting property is introduced to interpret temporary correlated fluctuations from an overall degrading trend in degradation records. The corresponding parameter estimation is proposed based on the maximum likelihood estimation method, and a Volterra integral equation of second kind with a non-singular kernel is then considered to calculate the probability density function (pdf) of failure time. Proposed methods are tested in a case study, where results are compared with a nonlinear-drift, linear-diffusion process.

International audience; In the literature, the effectiveness of a maintenance strategy is usually assessed through a cost criterion which is the long-run expected maintenance cost rate (i.e., performance viewpoint). However, such a criterion does not allow evaluating the variability of maintenance costs from a renewal cycle to another, and classical maintenance strategies seem inappropriate in the sense of risk management (i.e., robustness viewpoint). Therefore, this paper aims at (i) re-evaluating classical strategies from both performance and robustness aspects, and hence (ii) suggesting more suitable maintenance decisions. Especially, by defining the long-run expected maintenance cost rate as the performance criterion and the variance of maintenance cost per renewal cycle as the robustness criterion, we consider two representatives of time-based and condition-based maintenance families: a block replacement strategy, and a periodic inspection and replacement strategy. Their mathematical cost models are developed on the basis of the homogeneous Gamma degradation process and the theory of probability. The comparison results of both maintenance strategies show that the strategy which has a higher performance incurs a higher level of risk. So, it is necessary to assess jointly the performance and the robustness of maintenance strategies to find out a more reliable decision.

COM; International audience

International audience; In this paper, a novel predictive maintenance policy with multi-level decision-making is proposed for multi-component system with complex structure. The main idea is to propose a decision-making process considered on two levels: system level and component one. The goal of the decision rules at the system level is to address if preventive maintenance actions are needed regarding the predictive reliability of the system. At component level the decision rules aim at identifying optimally a group of several components to be prevntively maintained when preventive maintenance is trigged due to the system level decision. Selecting optimal components is based on a cost-based group improvement factor taking into account the predictive reliability of the components, the economic dependencies as well as the location of the components in the system. Moreover, a cost model is developed to find the optimal maintenance decision variables. A 14-component system is finally introduced to illustrate the use and the performance of the proposed predictive maintenance policy. Different sensitivity analysis are also investigated and discussed. Indeed, the proposed policy provides more flexibility in maintenance decision-making for complex structure systems, hence leading to significant profits in terms of maintenance cost when compared with existing policies.

International audience

International audience

International audience; In this paper, a modeling framework of a degrading control system is proposed with a focus on actuator deterioration. An actuator is costly and is the physical link between the control action and the governed process. For these reasons, it is a critical component. It is assumed a relationship between the controller setting and the actuator deterioration process. Moreover, as the degradation level increases, the actuator has more and more difficulties implementing the control action. The prognosis of the remaining useful lifetime of an actuator is then derived. Finally, a condition-based maintenance policy is proposed including the possibility to reconfigure the controller setting in order to be less stressful for the actuator. The impact of this additional action on the maintenance cost as well as on the system performances is assessed for several controller settings.

International audience

International audience; A major concern for engineers and managers nowadays is to make high quality products and highly reliable systems. In this context, reliability analysis and efficient maintenance decision-making are strongly required. Reliability analysis needs collecting, modeling and analysis different types of measure especially for deteriorating systems. In presence of information data on system deterioration level or health indicator, the development of an accurate mathematical model describing the system behavior is indispensable. Stochastic processes are suitable tools to model the system behavior by taking into account its dynamic evolution and the possible influencing random phenomenon. The stochastic processes are largely used in reliability modeling and prognostics e.g. Markov chains, birth processes, counting processes, Gamma processes, and Wiener processes. In the case of gradually deteriorating systems, mainly Gamma and Wiener processes are considered to model the evolution of the deterioration in time respectively for monotonically deteriorating systems and for non-monotone phenomenon. A more general framework that encompasses these latter models is the diffusion process. In this framework the process is defined as the solution of a stochastic differential equation which means that each infinitesimal increment of degradation is described by the sum of drift and a random perturbation. Sudden events and relatively large fluctuations that may appear in real data variation can also be represented by allowing the stochastic process to include “jumps factors”. One important type of stochastic process that has the ability to describe the general behavior of data evolution as well as the particular large fluctuations is the jump-diffusion process. In this paper, based on a case study related to real collected data, the aim is to predict as precise as possible the failure time presented by the first crossing time of a given failure threshold by the deterioration level. In this purpose, a jump-diffusion process is used to model the data evolution and predict the future behavior of the system under consideration. The given dataset consists in several independent nonmonotone trajectories considered as the deterioration trajectories of different items. The variance behavior is found to be similar for the different independent trajectories. According to the variance behavior a division of the time do main into three different states is proposed. The transition between these states is supposed to be controlled by an explanatory variable, called covariate, following a Markov chain. At each state, data evolution is considered to follow the jump-diffusion process with Log-Uniform Jump Amplitude (log-normal distributed increments with jumps). The data fitting and parameters estimations are detailed in the rest of this paper. Eventually, the resulting model is used in forecasting and risk prediction. Giving an estimation of the future evolution of system within a certain confidence interval, this model permits to schedule the maintenance actions.

International audience; This paper addresses the case of automatic controlled system which deteriorates during its operation because of components’ wear or deterioration. Depending on its specific closed-loop structure, the controlled system has the ability to compensate for disturbances affecting the actuators which can remain partially hidden. The deterioration modeling and the Remaining Useful Lifetime (RUL) estimation for such closed-loop dynamic system have not been addressed extensively. In this paper, we consider a controlled system with Proportional-Integral-Derivative controller. It is assumed that the actuator is subject to shocks that occur randomly in time. An integrated model is proposed to jointly describe the state of the controlled process and the actuator deterioration. Only the output of the controlled system is available to assess its health condition. By considering a Piecewise Deterministic Markov Process, the RUL of the system can be estimated by a two-step approach. In the first step referred as the “Diagnosis” step, the system state is estimated online from the available monitoring observations by using a particle filtering method. In the second step referred as the “Prognosis” step, the RUL is estimated as a conditional reliability by Monte Carlo simulation. To illustrate the approach, a simulated tank level control system is used.

ACL; International audience; The aim of this article is twofold: (i) modeling partially observed crack growth of industrial components using gamma processes and (ii) providing estimators of the best maintenance time in a statistical Bayesian framework. The choice of a Bayesian framework is motivated by the small size of data, the availability of expert knowledge about the crack propagation, and more generally, the concern about the integration of parametrical uncertainties when optimizing a maintenance action. The article answers to the methodological question of Bayesian prior elicitation by adopting a strategy based on virtual data information and defines optimal replacement times as posterior Bayes estimators minimizing appropriate cost functions. More precisely, the industrial data are described, and two different levels of available information are considered. Then, the Bayesian parameter estimation procedure in each case is thoroughly explained, by conducting MCMC runs. Different criteria for maintenance optimization, taking account all uncertainties, are considered and discussed. The overall procedure is tested on simulated data and applied over a real dataset. Copyright © 2014 John Wiley & Sons, Ltd.

International audience; The aim of this paper is to model degradation phenomena in a multi-unit context , with dependence between components induced by a common and external stressing random environment. The stress arrives by shocks and each shock induces a sudden increase to the deterioration level of each component. The model induces a two-step dependence between components: …rstly, all components are simultaneously impacted by a shock; secondly, for a given shock, the deterioration increments of the di¤erent components are correlated. This leads to a multivariate degradation model. A method of moments is proposed, for the estimation of the model parameters. Assuming the components to belong to a system with a given failure zone, the system reliability is next provided and the in ‡uence of the model parameters on the system lifetime is studied. Numerical experiments illustrate the study.

International audience

International audience; The phenomenon of threshold voltage instability has been reported to be a critical issue for the first generation of commercial SiC MOSFETs. For aeronautic mission profile, the long term reliability of such device has to be estimated prior to any integration. As far as the author knows, this paper proposes one of the first attempts to use a non-homogeneous gamma process to model the threshold voltage degradation of a commercial SiC MOSFET. Such approach allows evaluating the time to failure as well as it distribution thanks to the proposed Acceleration degradation testing methodology. All the data collected during the ageing tests will be used to model an acceleration factor and to extrapolate the Time-To-Failure in a given application. Ultimately, the use of gamma process to evaluate the Remaining Useful Life of electronic component is described as a first step toward prognostic of Remaining Useful Life in embedded power electronics.

International audience; In this paper, the maintenance of a system which undergoes a change in its deterioration rate is considered. The time of change of regime and the parameters after the change are unknown. To deal with unknown change time and unknown parameters, a detection procedure based on a suitable online change detection algorithm is used. The paper proposes a maintenance decision rule versus detection policy in order to minimise the long-run average maintenance cost, and the performances of this policy are studied through numerical implementations.

International audience; This paper aims at analyzing the effectiveness of some well-known condition indices for condition-based maintenance decision-making. The study is based on a single-unit system subject to competing risks due to multiple degradation paths of stress corrosion cracking (SCC). The system is assumed to be failed whenever the number of cracks or the sum of all crack sizes reach critical thresholds. A periodic inspection scheme is used to monitor the system health state. Each inspection returns the perfect diagnostic information (i.e., the number of cracks and the corresponding crack size, as well as the failure/working state of the system). Based on this information, two kinds of prognostic condition indices (i.e., average type and quantile type of the system residual useful life) are constructed. The associated predictive maintenance strategies and cost models are also developed and compared with the ones whose maintenance decisions are based on diagnostic condition indices. The comparison results allow us to conclude on the performance and on the robustness of the prognostic indices compared to the diagnostic ones, hence giving some suggestions on the choice of condition indices for maintenance decision-making.

International audience; This paper deals with a predictive maintenance strategy for multi-component systems whose structure may impact the evolution of components' deterioration that can be inspected at regular intervals of time. A decision rule relied on both structural importance of components and their predictive reliability is proposed to select components to be preventively maintained. The economic dependencies between components that can save maintenance costs are also investigated. Moreover, for corrective maintenance actions, an intelligent opportunistic rule taking into account both importance of components and logistic support constraints is proposed. A numerical example of five-component system is finally introduced to illustrate the proposed maintenance strategy.

11 pages; International audience; This paper deals with a gradually deteriorating system operating under an uncertain environment whose state is only known on a finite rolling horizon. As such, the system is subject to constraints. Maintenance actions can only be planned at imposed times called maintenance opportunities that are available on a limited visibility horizon.This system can, for example, be a commercial vehicle with a monitored critical component that can be maintained only in some specific workshops. Based on the considered system, we aim to use the monitoring data and the time-limited information for maintenance decision support in order to reduce its costs. We propose two predictive maintenance policies based, respectively, on cost and reliability criteria. Classical age-based and condition-based policies are considered as benchmarks. The performance assessment shows the value of the different types of information and the best way to use them in maintenance decision making.

International audience; The aim of this paper is to propose an efficient condition-based maintenance policy for a gradually deteriorating system. In this purpose a stochastic deterioration model is proposed according to the feedback data and once the deterioration is modelled a condition-based maintenance policy is proposed which minimises the long run mean average cost. The performances of the proposed condition-based maintenance policy are evaluated through Monte-Carlo simulation methods. To illustrate results Stress Corrosion Cracking phenomenon is considered.

International audience; Developments of monitoring techniques provide information on the actual conditions. The information level available on the system (failure times, deterioration, load, usage condition, etc.) and the way to integrate them impact directly the performance of maintenance operations that represent a substantial portion of the total life cycle costs of many systems. In this context, we consider a gradually deteriorating system operating under an uncertain environment. The information about the future environment state is only known on a finite rolling horizon. The system is subject to constraints and maintenance actions cannot be planned at any time, but at fixed times called maintenance opportunities and known only on a finite horizon. Based on the considered system, we aim to use the monitoring data and the time-limited information for maintenance decision support in order to reduce its costs. Several maintenance policies are proposed: age-based policy, condition-based policy and two predictive policies based on a cost and a risk criterions. The comparison of the maintenance cost savings of these policies allows concluding the value of different types of information and the best ways to use them in maintenance decision-making

Chapter 8; International audience

The aim of this paper is to combine the deterministic behavior of a feedback control system with an inner stochastic degradation process. For that, a modeling framework is proposed with an intrinsic deterioration caused by the control input law and its variations. As a consequence, when degrading the actuator has a loss of effectiveness to carry out this command. Two system missions are studied - slow and fast missions - with a Markov chain allowing to parameterize the sojourn times in the different system states. Finally, a basic PID controler commonly used in industrial automation is assessed regarding performances and lifetime aided with Monte Carlo simulations.

ACT; International audience; no abstract

International audience; This paper deals with a gradually deteriorating system which can be for example a large structure such as a bridge, a dike or a pipeline. A such system should be preventively maintained in order to avoid failures which would have disastrous consequences. A condition-based maintenance policy is then proposed considering a finite horizon. As preventive maintenance actions are assumed to be imperfect their impact on the system is determined through a monotone decreasing improvement function. The aim of this paper is to propose an inspection interval in order to numerically assess, on a finite time span, a condition-based maintenance policy.

International audience; The presented system monitors a turbofan start sequence using indicators and operating conditions to detect abnormal behavior. It is based on the analysis of the residuals between the measured indicator values and the corresponding estimated values assuming healthy state. Estimation uses regression models trained on a database. However, as in many monitoring problems, the amount of data is limited due to application issues and covers only a limited region of the feature space. Thus, the models are trained in a limited domain defined implicitly by the available learning data and their efficiency is not controlled outside this implicit domain. This paper deals with the definition and the extension of the models validity region while keeping the extension effect on the monitoring process under control. A methodology based on one-class SVM is proposed and is applied to the presented monitoring system. Practical and methodological conclusions are drawn from the proposed experiments.

International audience

International audience; This paper considers the age-based maintenance of a two-identical components system with economic dependences. Two existing approaches are presented and compared. One is a maintenance policy of (N,n)-type. It is very close to the optimal policy among all possible policies but difficult to optimize: Policy Iteration Algorithm is needed. The aim of the paper is to investigate if there are cases for which a simpler policy than those of (N,n)-type can be sufficient in terms of maintenance cost. A second approach proposed in literature (derived from the time-based block replacement one) is simpler but not sufficient in any situations. Some numerical experiments and an analysis of the stochastic behavior of the two-identical components system allows us to propose a third policy (Policy P). This third maintenance plan is simplified from the second one. It compensates for its weak point which is to be very expensive when the cost of replacement for one component is close to the cost of replacement for both components. Moreover it is possible to generalize this third policy from a two-identical components system to a n-identical components system. The performances analysis of the policy is based on numerical experiments.

International audience

International audience; In this paper, a modelling approach is presented to assess the performance of multicomponent systems maintained by complex maintenance strategies. A global framework is built to describe the evolution of the system, the way its components can degrade and fail, and the effects of different maintenance actions. A two-level modelling framework makes it possible to describe fully the entire causal chain that can lead to system dysfunction and the possible maintenance tasks of the reliability centred maintenance (RCM) method. A cost model is defined to assess system performance in terms of cost and unavailability. The main originality of this work is that the complexity of both maintenance programmes and system structures is taken into account. The model can be used as a decision-making tool to choose from a selection of various maintenance options. Promising results have been obtained in case studies based on systems from nuclear power plants.