Imene Djelloul - Academia.edu (original) (raw)
Papers by Imene Djelloul
Proceedings Of The Institution Of Mechanical Engineers, Part I: Journal Of Systems And Control Engineering, Jan 16, 2019
Fault diagnosis is becoming an important issue in industrial environment, and the accurate diagno... more Fault diagnosis is becoming an important issue in industrial environment, and the accurate diagnosis is the most significant part in fault handling. This article discusses a fault detection and diagnosis problem for manufacturing systems taking into account rapid detection and speed performances of fault isolation with minimum ambiguity. However, in many complex real plants, it may not be possible to discover accurately the causes of probable faults. The accuracy of fault or fault detection by the traditional approaches is not adequate. Considering the quality effect of the learning algorithm, a new hybrid neural network approach is developed using the integration of a regression task for classification accuracy. Two models of neural networks: gradient descent and momentum & adaptive learning rate and Levenberg-Marquardt are investigated and compared. The performance of the proposed approach is evaluated using mean square error, convergence speed, and classification accuracy. The case study and experimental results are presented and discussed. A comparison with the Levenberg-Marquardt regression approach shows the importance of considering the proposed learning algorithm quality in the fault detection and diagnosis problem compared with those reported in the literature.
Neural Computing and Applications, Aug 30, 2017
This paper addresses the fault detection and isolation problem in manufacturing systems. Some of ... more This paper addresses the fault detection and isolation problem in manufacturing systems. Some of these systems can be affected by several faults, a first way of determining them is to use classification and rule-based reasoning methods. In the present work, a new hybrid algorithm, based on fuzzy Levenberg-Marquardt and genetic algorithm for both training and fault isolation in the high-dimensional setting, is developed. The genetic algorithm-based approach aims at selecting an optimal number of production rules. The developed approach consists then to minimize training time and to find accurate and interpretable fuzzy systems with an appropriate production rules subset. It is put into practice for a real manufacturing system for binary classes and also its advantage is demonstrated on multi-class system. Obtained results show that the approach can be more accurate and fast to make fault diagnosis for binary and multi-class problems compared to those reported in the literature.
Applied Intelligence, Feb 8, 2018
This paper is concerned with fault detection and diagnosis problem in manufacturing systems. In s... more This paper is concerned with fault detection and diagnosis problem in manufacturing systems. In such industrial environment, production systems are subject to several faults caused by a number of factors including the environment, the accumulated wearing, usage, etc. However, due to the lack of accuracy or fluctuation of data, it is oftentimes impossible to evaluate precisely the correct classification rate of faults. In order to classify each type of fault, neural networks and fuzzy logic are two different intelligent diagnosis methods that are more applied now, and each has its own advantages and disadvantages. A new hybrid fault diagnosis approach is introduced in this paper that considers the combined learning algorithm and knowledge base (Fuzzy rules) to handle ambiguous and even erroneous information. Therefore, to enhance the classification accuracy, three perceptron models including: linear perceptron (LP), multilayer perceptron (MLP) and fuzzy perceptron (FP) have been respectively established and compared. The conditional risk function "PDF" that measures the expectation of loss when taking an action is presented at the same time. We evaluate the proposed hybrid approach "Variable Learning Rate Gradient Descent with Bayes' Maximum Likelihood formula" VLRGD-BML on dataset of milk pasteurization process and compare our approach with other similar published works for fault diagnosis in the literature. Comparative results indicate the higher efficiency and effectiveness of the proposed approach with fuzzy perceptron for uncertain fault diagnosis problem.
HAL (Le Centre pour la Communication Scientifique Directe), Dec 2, 2015
Ce papier présente une stratégie de maintenance sélective pour un système multi-composants requis... more Ce papier présente une stratégie de maintenance sélective pour un système multi-composants requis pour effectuer une série de missions avec des arrêts finis planifiés entre deux missions successives. Pendant ces arrêts, les actions de maintenance peuvent être réalisées sur certains composants du système. Pour chaque composant, une liste d'actions de maintenance est disponible où figurent les actions de maintenance parfaite, minimale ou imparfaite. Tenant compte des limitations sur les ressources en maintenance telles que le temps et le budget, il est parfois impossible d'effectuer toutes les opérations de maintenance désirées. Le problème de la maintenance sélective vise donc à sélectionner les composants qui doivent être maintenus afin de maximiser la fiabilité du système pour exécuter la prochaine mission. Dans ce travail, les durées des missions sont considérées aléatoires et représentées par des variables aléatoires. Un modèle d'optimisation mathématique de la maintenance sélective est ensuite proposé et dont l'objectif est de maximiser la fiabilité du système à exécuter sa prochaine mission, en tenant compte des contraintes de budget et du temps alloués à la maintenance. L'intérêt de notre approche est démontré sur un exemple de système séries-parallèle.
Journal of Manufacturing Systems, Apr 1, 2017
This paper deals with selective maintenance of a multi-component system, performing several missi... more This paper deals with selective maintenance of a multi-component system, performing several missions seceded by scheduled breaks. To improve the probability of successfully completing its next mission, the system's components are maintained during these breaks. A list of possible maintenance actions on each component of the system, ranging from minimal repair to overhaul through intermediate imperfect maintenance actions, is available. Durations of the missions as well as those of the breaks are assumed stochastic variables with known distributions. The resulting selective maintenance optimization problem is thus modeled as a mixed-integer non-linear stochastic program. Its objective is to determine a costoptimal subset of maintenance actions to be performed on the system's components, during the limited stochastic duration of the break, to meet a predetermined minimum system's reliability to operate the next mission. The fundamental constructs and the relevant parameters of this decision-making problem are developed and discussed. An illustrative example is provided to demonstrate the added value of solving this selective maintenance problem as a stochastic optimization program.
3rd International Conference on Systems and Control, 2013
ABSTRACT In this paper, our interest is focused on monitoring in production systems. The motivati... more ABSTRACT In this paper, our interest is focused on monitoring in production systems. The motivation behind this investigation is the need of presenting hybrid approach of Neuro-Fuzzy Genetic for the optimal learning, which allows getting the required performance measures. The presence of Neuro- Fuzzy algorithms which may involving elegantly the set of features extraction are defined in terms of membership function, where as Genetic Algorithms are proposed to optimize the set of rules, in order to design supervised classification systems by generating fuzzy if-then rules. The learning process is based on two algorithms: Levenberg-Marquardt (TRAINLM) and Gradient Descent (TRAINGDA). Then, the proposed algorithms performances are verified and analyzed through an industrial application of agro-alimentary unit called AURES dairy for the city of Batna. The simulation results show that the proposed approach performs the best when Levenberg-Marquardt is used as a learning algorithm.
Fault is one of the main causes of failure, and the accurate diagnosis is one of the most signifi... more Fault is one of the main causes of failure, and the accurate diagnosis is one of the most significant steps in fault treatment. This paper considers the diagnosis system to solve some maintenance optimization problems in manufacturing systems. The proposed architecture deals primarily with three modules, namely, the detection module, the diagnosis module, and the decision making module. In this case, the fault needs to be detected and diagnosed as early as possible after its occurrence. Data mining techniques can support repairmen in diagnosis decision-making process. To be successful, we suggest new classification approach based on hybrid neural network technique focusing this industrial application for developing a diagnosis system. Two models of neural networks: Gradient Descent and Momentum & Adaptive LR and Levenberg-Marquardt are investigated. Classifier system was used in order to construct accurate system for fault classification based on regression technique. The performance of the approach is evaluated using mean square error and classification accuracy. Case study and experimental results are given and discussed. Results achieved in this paper have potential to open new opportunities in industrial diagnosis of probable faults.
Le Centre pour la Communication Scientifique Directe - HAL - Université Paris Descartes, Dec 2, 2015
Ce papier présente une stratégie de maintenance sélective pour un système multi-composants requis... more Ce papier présente une stratégie de maintenance sélective pour un système multi-composants requis pour effectuer une série de missions avec des arrêts finis planifiés entre deux missions successives. Pendant ces arrêts, les actions de maintenance peuvent être réalisées sur certains composants du système. Pour chaque composant, une liste d'actions de maintenance est disponible où figurent les actions de maintenance parfaite, minimale ou imparfaite. Tenant compte des limitations sur les ressources en maintenance telles que le temps et le budget, il est parfois impossible d'effectuer toutes les opérations de maintenance désirées. Le problème de la maintenance sélective vise donc à sélectionner les composants qui doivent être maintenus afin de maximiser la fiabilité du système pour exécuter la prochaine mission. Dans ce travail, les durées des missions sont considérées aléatoires et représentées par des variables aléatoires. Un modèle d'optimisation mathématique de la maintenance sélective est ensuite proposé et dont l'objectif est de maximiser la fiabilité du système à exécuter sa prochaine mission, en tenant compte des contraintes de budget et du temps alloués à la maintenance. L'intérêt de notre approche est démontré sur un exemple de système séries-parallèle.
International Journal of Production Research, 2017
This paper deals with the selective maintenance problem for a multi-component system performing c... more This paper deals with the selective maintenance problem for a multi-component system performing consecutive missions separated by scheduled breaks. To increase the probability of successfully completing its next mission, the system components are maintained during the break. A list of potential imperfect maintenance actions on each component, ranging from minimal repair to replacement is available. The general hybrid hazard rate approach is used to model the reliability improvement of the system components. Durations of the maintenance actions, the mission and the breaks are stochastic with known probability distributions. The resulting optimisation problem is modelled as a non-linear stochastic programme. Its objective is to determine a cost-optimal subset of maintenance actions to be performed on the components given the limited stochastic duration of the break and the minimum system reliability level required to complete the next mission. The fundamental concepts and relevant parameters of this decision-making problem are developed and discussed. Numerical experiments are provided to demonstrate the added value of solving this selective maintenance problem as a stochastic optimisation programme.
This paper deals with selective maintenance of systems required to perform missions with finite b... more This paper deals with selective maintenance of systems required to perform missions with finite breaks between consecutive missions. During breaks, maintenance actions may be performed on system's components. For each component, a list of maintenance actions is available, from minimal repair to replacement, trough imperfect maintenance. Due to limitations on maintenance resources such as time and budget, it may not be possible to perform all necessary maintenance actions. The selective maintenance problem consists then to select a subset of maintenance actions to be performed on some components to maximize the next mission reliability. In this paper, the selective maintenance problem is investigated for a series-parallel system. Missions operated by the system are considered of uncertain durations and modeled as random variables. A mathematical optimization model is then proposed where the objective is to maximize the reliability of executing the next mission, taking into accoun...
This chapter addresses a maintenance optimization problem for re-manufactured equipments that wil... more This chapter addresses a maintenance optimization problem for re-manufactured equipments that will be reintroduced into the market as second-hand equipments. The main difference of this work and the previous literature on the maintenance optimization of second-hand equipments is the influence of the uncertainties due to the indirect obsolescence concept. The uncertainty is herein about the spare parts availability to perform some maintenance actions on equipment due to technology vanishing. The maintenance policy involves in fact a minimal repair at failure and a preventive repair after some operating period. To deal with this shortcoming, the life cycle of technology or spare parts availability is defined and modeled as a random variable whose lifetimes distribution is well known and Weibull distributed. Accordingly, an optimal maintenance policy is discussed and derived for such equipment in order to overcome the uncertainty on reparation action. Moreover, experiments are then con...
Fault is one of the main causes of failure, and the accurate diagnosis is one of the most signifi... more Fault is one of the main causes of failure, and the accurate diagnosis is one of the most significant steps in fault treatment. This paper considers the diagnosis system to solve some maintenance optimization problems in manufacturing systems. The proposed architecture deals primarily with three modules, namely, the detection module, the diagnosis module, and the decision making module. In this case, the fault needs to be detected and diagnosed as early as possible after its occurrence. Data mining techniques can support repairmen in diagnosis decision-making process. To be successful, we suggest new classification approach based on hybrid neural network technique focusing this industrial application for developing a diagnosis system. Two models of neural networks: Gradient Descent and Momentum & Adaptive LR and Levenberg-Marquardt are investigated. Classifier system was used in order to construct accurate system for fault classification based on regression technique. The performanc...
This paper deals with selective maintenance of a multi-component system, performing several missi... more This paper deals with selective maintenance of a multi-component system, performing several missions seceded by scheduled breaks. To improve the probability of successfully completing its next mission, the system's components are maintained during these breaks. A list of possible maintenance actions on each component of the system, ranging from minimal repair to overhaul through intermediate imperfect maintenance actions, is available. Durations of the missions as well as those of the breaks are assumed stochastic variables with known distributions. The resulting selective maintenance optimization problem is thus modeled as a mixed-integer non-linear stochastic program. Its objective is to determine a cost-optimal subset of maintenance actions to be performed on the system's components, during the limited stochastic duration of the break, to meet a predetermined minimum system's reliability to operate the next mission. The fundamental constructs and the relevant paramete...
IFAC-PapersOnLine
This paper addresses the selective maintenance optimization problem for a series-parallel system.... more This paper addresses the selective maintenance optimization problem for a series-parallel system. The system performs several missions with breaks between each two consecutive missions. To improve the system reliability to operate the next mission, its components are maintained during breaks. To each component, a list of maintenance actions is available from minimal repair to overhaul through imperfect maintenance actions. Durations of missions as well as that of breaks are considered not constant but rather stochastic. These durations are therefore modeled as random variables. The selective maintenance optimization problem developed is then non-linear and stochastic. The fundamental constructs and the relevant parameters of this decision-making problem are developed and discussed. A simplified series-parallel system is investigated to demonstrate the added value of solving this non-linear and stochastic selective maintenance optimization program.
System Reliability, Dec 20, 2017
This chapter investigates optimization of maintenance policy of a repairable equipment whose life... more This chapter investigates optimization of maintenance policy of a repairable equipment whose lifetime distribution depends on the operating environment severity. The considered equipment is undergone to a maintenance policy which consists of repairing minimally at failure and maintaining after operating periods. The periodic maintenance is preventive maintenance (PM) and allows reducing consequently the equipment age but with higher cost than minimal repair. In addition, the equipment has to operate at least in two operating environments with different severity. Therefore, in this analysis, the equipment lifetime distribution function depends on the operating severity. Under these hypotheses, a mathematical modeling of the maintenance cost per unit of time is proposed and discussed. This cost is mathematically analyzed in order to derive optimal periods between preventive maintenance (PM) and the optimal condition under which these exist.
Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering
Fault diagnosis is becoming an important issue in industrial environment, and the accurate diagno... more Fault diagnosis is becoming an important issue in industrial environment, and the accurate diagnosis is the most significant part in fault handling. This article discusses a fault detection and diagnosis problem for manufacturing systems taking into account rapid detection and speed performances of fault isolation with minimum ambiguity. However, in many complex real plants, it may not be possible to discover accurately the causes of probable faults. The accuracy of fault or fault detection by the traditional approaches is not adequate. Considering the quality effect of the learning algorithm, a new hybrid neural network approach is developed using the integration of a regression task for classification accuracy. Two models of neural networks: gradient descent and momentum & adaptive learning rate and Levenberg–Marquardt are investigated and compared. The performance of the proposed approach is evaluated using mean square error, convergence speed, and classification accuracy. The ca...
Neural Computing and Applications
Proceedings Of The Institution Of Mechanical Engineers, Part I: Journal Of Systems And Control Engineering, Jan 16, 2019
Fault diagnosis is becoming an important issue in industrial environment, and the accurate diagno... more Fault diagnosis is becoming an important issue in industrial environment, and the accurate diagnosis is the most significant part in fault handling. This article discusses a fault detection and diagnosis problem for manufacturing systems taking into account rapid detection and speed performances of fault isolation with minimum ambiguity. However, in many complex real plants, it may not be possible to discover accurately the causes of probable faults. The accuracy of fault or fault detection by the traditional approaches is not adequate. Considering the quality effect of the learning algorithm, a new hybrid neural network approach is developed using the integration of a regression task for classification accuracy. Two models of neural networks: gradient descent and momentum & adaptive learning rate and Levenberg-Marquardt are investigated and compared. The performance of the proposed approach is evaluated using mean square error, convergence speed, and classification accuracy. The case study and experimental results are presented and discussed. A comparison with the Levenberg-Marquardt regression approach shows the importance of considering the proposed learning algorithm quality in the fault detection and diagnosis problem compared with those reported in the literature.
Neural Computing and Applications, Aug 30, 2017
This paper addresses the fault detection and isolation problem in manufacturing systems. Some of ... more This paper addresses the fault detection and isolation problem in manufacturing systems. Some of these systems can be affected by several faults, a first way of determining them is to use classification and rule-based reasoning methods. In the present work, a new hybrid algorithm, based on fuzzy Levenberg-Marquardt and genetic algorithm for both training and fault isolation in the high-dimensional setting, is developed. The genetic algorithm-based approach aims at selecting an optimal number of production rules. The developed approach consists then to minimize training time and to find accurate and interpretable fuzzy systems with an appropriate production rules subset. It is put into practice for a real manufacturing system for binary classes and also its advantage is demonstrated on multi-class system. Obtained results show that the approach can be more accurate and fast to make fault diagnosis for binary and multi-class problems compared to those reported in the literature.
Applied Intelligence, Feb 8, 2018
This paper is concerned with fault detection and diagnosis problem in manufacturing systems. In s... more This paper is concerned with fault detection and diagnosis problem in manufacturing systems. In such industrial environment, production systems are subject to several faults caused by a number of factors including the environment, the accumulated wearing, usage, etc. However, due to the lack of accuracy or fluctuation of data, it is oftentimes impossible to evaluate precisely the correct classification rate of faults. In order to classify each type of fault, neural networks and fuzzy logic are two different intelligent diagnosis methods that are more applied now, and each has its own advantages and disadvantages. A new hybrid fault diagnosis approach is introduced in this paper that considers the combined learning algorithm and knowledge base (Fuzzy rules) to handle ambiguous and even erroneous information. Therefore, to enhance the classification accuracy, three perceptron models including: linear perceptron (LP), multilayer perceptron (MLP) and fuzzy perceptron (FP) have been respectively established and compared. The conditional risk function "PDF" that measures the expectation of loss when taking an action is presented at the same time. We evaluate the proposed hybrid approach "Variable Learning Rate Gradient Descent with Bayes' Maximum Likelihood formula" VLRGD-BML on dataset of milk pasteurization process and compare our approach with other similar published works for fault diagnosis in the literature. Comparative results indicate the higher efficiency and effectiveness of the proposed approach with fuzzy perceptron for uncertain fault diagnosis problem.
HAL (Le Centre pour la Communication Scientifique Directe), Dec 2, 2015
Ce papier présente une stratégie de maintenance sélective pour un système multi-composants requis... more Ce papier présente une stratégie de maintenance sélective pour un système multi-composants requis pour effectuer une série de missions avec des arrêts finis planifiés entre deux missions successives. Pendant ces arrêts, les actions de maintenance peuvent être réalisées sur certains composants du système. Pour chaque composant, une liste d'actions de maintenance est disponible où figurent les actions de maintenance parfaite, minimale ou imparfaite. Tenant compte des limitations sur les ressources en maintenance telles que le temps et le budget, il est parfois impossible d'effectuer toutes les opérations de maintenance désirées. Le problème de la maintenance sélective vise donc à sélectionner les composants qui doivent être maintenus afin de maximiser la fiabilité du système pour exécuter la prochaine mission. Dans ce travail, les durées des missions sont considérées aléatoires et représentées par des variables aléatoires. Un modèle d'optimisation mathématique de la maintenance sélective est ensuite proposé et dont l'objectif est de maximiser la fiabilité du système à exécuter sa prochaine mission, en tenant compte des contraintes de budget et du temps alloués à la maintenance. L'intérêt de notre approche est démontré sur un exemple de système séries-parallèle.
Journal of Manufacturing Systems, Apr 1, 2017
This paper deals with selective maintenance of a multi-component system, performing several missi... more This paper deals with selective maintenance of a multi-component system, performing several missions seceded by scheduled breaks. To improve the probability of successfully completing its next mission, the system's components are maintained during these breaks. A list of possible maintenance actions on each component of the system, ranging from minimal repair to overhaul through intermediate imperfect maintenance actions, is available. Durations of the missions as well as those of the breaks are assumed stochastic variables with known distributions. The resulting selective maintenance optimization problem is thus modeled as a mixed-integer non-linear stochastic program. Its objective is to determine a costoptimal subset of maintenance actions to be performed on the system's components, during the limited stochastic duration of the break, to meet a predetermined minimum system's reliability to operate the next mission. The fundamental constructs and the relevant parameters of this decision-making problem are developed and discussed. An illustrative example is provided to demonstrate the added value of solving this selective maintenance problem as a stochastic optimization program.
3rd International Conference on Systems and Control, 2013
ABSTRACT In this paper, our interest is focused on monitoring in production systems. The motivati... more ABSTRACT In this paper, our interest is focused on monitoring in production systems. The motivation behind this investigation is the need of presenting hybrid approach of Neuro-Fuzzy Genetic for the optimal learning, which allows getting the required performance measures. The presence of Neuro- Fuzzy algorithms which may involving elegantly the set of features extraction are defined in terms of membership function, where as Genetic Algorithms are proposed to optimize the set of rules, in order to design supervised classification systems by generating fuzzy if-then rules. The learning process is based on two algorithms: Levenberg-Marquardt (TRAINLM) and Gradient Descent (TRAINGDA). Then, the proposed algorithms performances are verified and analyzed through an industrial application of agro-alimentary unit called AURES dairy for the city of Batna. The simulation results show that the proposed approach performs the best when Levenberg-Marquardt is used as a learning algorithm.
Fault is one of the main causes of failure, and the accurate diagnosis is one of the most signifi... more Fault is one of the main causes of failure, and the accurate diagnosis is one of the most significant steps in fault treatment. This paper considers the diagnosis system to solve some maintenance optimization problems in manufacturing systems. The proposed architecture deals primarily with three modules, namely, the detection module, the diagnosis module, and the decision making module. In this case, the fault needs to be detected and diagnosed as early as possible after its occurrence. Data mining techniques can support repairmen in diagnosis decision-making process. To be successful, we suggest new classification approach based on hybrid neural network technique focusing this industrial application for developing a diagnosis system. Two models of neural networks: Gradient Descent and Momentum & Adaptive LR and Levenberg-Marquardt are investigated. Classifier system was used in order to construct accurate system for fault classification based on regression technique. The performance of the approach is evaluated using mean square error and classification accuracy. Case study and experimental results are given and discussed. Results achieved in this paper have potential to open new opportunities in industrial diagnosis of probable faults.
Le Centre pour la Communication Scientifique Directe - HAL - Université Paris Descartes, Dec 2, 2015
Ce papier présente une stratégie de maintenance sélective pour un système multi-composants requis... more Ce papier présente une stratégie de maintenance sélective pour un système multi-composants requis pour effectuer une série de missions avec des arrêts finis planifiés entre deux missions successives. Pendant ces arrêts, les actions de maintenance peuvent être réalisées sur certains composants du système. Pour chaque composant, une liste d'actions de maintenance est disponible où figurent les actions de maintenance parfaite, minimale ou imparfaite. Tenant compte des limitations sur les ressources en maintenance telles que le temps et le budget, il est parfois impossible d'effectuer toutes les opérations de maintenance désirées. Le problème de la maintenance sélective vise donc à sélectionner les composants qui doivent être maintenus afin de maximiser la fiabilité du système pour exécuter la prochaine mission. Dans ce travail, les durées des missions sont considérées aléatoires et représentées par des variables aléatoires. Un modèle d'optimisation mathématique de la maintenance sélective est ensuite proposé et dont l'objectif est de maximiser la fiabilité du système à exécuter sa prochaine mission, en tenant compte des contraintes de budget et du temps alloués à la maintenance. L'intérêt de notre approche est démontré sur un exemple de système séries-parallèle.
International Journal of Production Research, 2017
This paper deals with the selective maintenance problem for a multi-component system performing c... more This paper deals with the selective maintenance problem for a multi-component system performing consecutive missions separated by scheduled breaks. To increase the probability of successfully completing its next mission, the system components are maintained during the break. A list of potential imperfect maintenance actions on each component, ranging from minimal repair to replacement is available. The general hybrid hazard rate approach is used to model the reliability improvement of the system components. Durations of the maintenance actions, the mission and the breaks are stochastic with known probability distributions. The resulting optimisation problem is modelled as a non-linear stochastic programme. Its objective is to determine a cost-optimal subset of maintenance actions to be performed on the components given the limited stochastic duration of the break and the minimum system reliability level required to complete the next mission. The fundamental concepts and relevant parameters of this decision-making problem are developed and discussed. Numerical experiments are provided to demonstrate the added value of solving this selective maintenance problem as a stochastic optimisation programme.
This paper deals with selective maintenance of systems required to perform missions with finite b... more This paper deals with selective maintenance of systems required to perform missions with finite breaks between consecutive missions. During breaks, maintenance actions may be performed on system's components. For each component, a list of maintenance actions is available, from minimal repair to replacement, trough imperfect maintenance. Due to limitations on maintenance resources such as time and budget, it may not be possible to perform all necessary maintenance actions. The selective maintenance problem consists then to select a subset of maintenance actions to be performed on some components to maximize the next mission reliability. In this paper, the selective maintenance problem is investigated for a series-parallel system. Missions operated by the system are considered of uncertain durations and modeled as random variables. A mathematical optimization model is then proposed where the objective is to maximize the reliability of executing the next mission, taking into accoun...
This chapter addresses a maintenance optimization problem for re-manufactured equipments that wil... more This chapter addresses a maintenance optimization problem for re-manufactured equipments that will be reintroduced into the market as second-hand equipments. The main difference of this work and the previous literature on the maintenance optimization of second-hand equipments is the influence of the uncertainties due to the indirect obsolescence concept. The uncertainty is herein about the spare parts availability to perform some maintenance actions on equipment due to technology vanishing. The maintenance policy involves in fact a minimal repair at failure and a preventive repair after some operating period. To deal with this shortcoming, the life cycle of technology or spare parts availability is defined and modeled as a random variable whose lifetimes distribution is well known and Weibull distributed. Accordingly, an optimal maintenance policy is discussed and derived for such equipment in order to overcome the uncertainty on reparation action. Moreover, experiments are then con...
Fault is one of the main causes of failure, and the accurate diagnosis is one of the most signifi... more Fault is one of the main causes of failure, and the accurate diagnosis is one of the most significant steps in fault treatment. This paper considers the diagnosis system to solve some maintenance optimization problems in manufacturing systems. The proposed architecture deals primarily with three modules, namely, the detection module, the diagnosis module, and the decision making module. In this case, the fault needs to be detected and diagnosed as early as possible after its occurrence. Data mining techniques can support repairmen in diagnosis decision-making process. To be successful, we suggest new classification approach based on hybrid neural network technique focusing this industrial application for developing a diagnosis system. Two models of neural networks: Gradient Descent and Momentum & Adaptive LR and Levenberg-Marquardt are investigated. Classifier system was used in order to construct accurate system for fault classification based on regression technique. The performanc...
This paper deals with selective maintenance of a multi-component system, performing several missi... more This paper deals with selective maintenance of a multi-component system, performing several missions seceded by scheduled breaks. To improve the probability of successfully completing its next mission, the system's components are maintained during these breaks. A list of possible maintenance actions on each component of the system, ranging from minimal repair to overhaul through intermediate imperfect maintenance actions, is available. Durations of the missions as well as those of the breaks are assumed stochastic variables with known distributions. The resulting selective maintenance optimization problem is thus modeled as a mixed-integer non-linear stochastic program. Its objective is to determine a cost-optimal subset of maintenance actions to be performed on the system's components, during the limited stochastic duration of the break, to meet a predetermined minimum system's reliability to operate the next mission. The fundamental constructs and the relevant paramete...
IFAC-PapersOnLine
This paper addresses the selective maintenance optimization problem for a series-parallel system.... more This paper addresses the selective maintenance optimization problem for a series-parallel system. The system performs several missions with breaks between each two consecutive missions. To improve the system reliability to operate the next mission, its components are maintained during breaks. To each component, a list of maintenance actions is available from minimal repair to overhaul through imperfect maintenance actions. Durations of missions as well as that of breaks are considered not constant but rather stochastic. These durations are therefore modeled as random variables. The selective maintenance optimization problem developed is then non-linear and stochastic. The fundamental constructs and the relevant parameters of this decision-making problem are developed and discussed. A simplified series-parallel system is investigated to demonstrate the added value of solving this non-linear and stochastic selective maintenance optimization program.
System Reliability, Dec 20, 2017
This chapter investigates optimization of maintenance policy of a repairable equipment whose life... more This chapter investigates optimization of maintenance policy of a repairable equipment whose lifetime distribution depends on the operating environment severity. The considered equipment is undergone to a maintenance policy which consists of repairing minimally at failure and maintaining after operating periods. The periodic maintenance is preventive maintenance (PM) and allows reducing consequently the equipment age but with higher cost than minimal repair. In addition, the equipment has to operate at least in two operating environments with different severity. Therefore, in this analysis, the equipment lifetime distribution function depends on the operating severity. Under these hypotheses, a mathematical modeling of the maintenance cost per unit of time is proposed and discussed. This cost is mathematically analyzed in order to derive optimal periods between preventive maintenance (PM) and the optimal condition under which these exist.
Proceedings of the Institution of Mechanical Engineers, Part I: Journal of Systems and Control Engineering
Fault diagnosis is becoming an important issue in industrial environment, and the accurate diagno... more Fault diagnosis is becoming an important issue in industrial environment, and the accurate diagnosis is the most significant part in fault handling. This article discusses a fault detection and diagnosis problem for manufacturing systems taking into account rapid detection and speed performances of fault isolation with minimum ambiguity. However, in many complex real plants, it may not be possible to discover accurately the causes of probable faults. The accuracy of fault or fault detection by the traditional approaches is not adequate. Considering the quality effect of the learning algorithm, a new hybrid neural network approach is developed using the integration of a regression task for classification accuracy. Two models of neural networks: gradient descent and momentum & adaptive learning rate and Levenberg–Marquardt are investigated and compared. The performance of the proposed approach is evaluated using mean square error, convergence speed, and classification accuracy. The ca...
Neural Computing and Applications