Search results

The purpose of this study is to propose and model an inspection and preventive maintenance policy for randomly failing systems that alternate operating and idle periods according to their mission profile.

A maintenance policy is defined and modeled mathematically. The paper focuses on finding the age T for inspection which maximizes the stationary availability of the system.

Except for the case of only self‐announcing failures, there always exists a finite optimal strategy T*. Two sufficient conditions for the uniqueness of such an optimum are also derived.

Many productive systems alternate operating and inactive periods, their failures may be self‐announcing or not self‐announcing (detected only through inspection). This paper presents a maintenance strategy for such systems in order to maximize their stationary availability. The proposed strategy suggests submitting the system to inspection when its age reaches T units of time.

This paper states a general expression of the system stationary availability which is considered as the performance criterion. Conditions of existence and uniqueness of an optimal strategy are developed.

The purpose of this paper is to determine a nearly optimal inspection sequence for a series system consisting of two components subject to gradual deterioration and whose failures are not self-announcing and can be detected only through inspection.

The problem is tackled in the context of condition-based maintenance (CBM) with a maintenance model in the class of the control-limit policies for which the maintenance decision is made following inspection by comparison of the deterioration level to critical thresholds. A mathematical model is developed to express the total expected cost per time unit as a function of the inspection instants.

For any given series system composed of two components with known critical deterioration threshold levels, and for any given set of costs related to inspection, inactivity due to failure, and preventive and corrective replacements of each component, a nearly optimal inspection sequence of the system is derived such as the total expected cost is reduced.

Due to the complexity of the cost model with the inspection instants (×1, ×2, ×3, …) being the decision variables, it has not been possible to derive the optimal solution. A quasi-optimal sequence of inspection times is derived along with the corresponding total average cost per time unit.

In many practical situations in which CBM is implemented, a tradeoff between inspection costs and inactivity and replacement costs has to be balanced when determining the intervals between successive inspections at which the degradation level of the components should be assessed and compared to predetermined critical threshold levels. Inspecting too often would increase inspection costs but in the same time it would also increase the probability to avoid a failure and end up with a preventive replacement, whereas not inspecting often enough would increase the probability to end up with a failure increasing replacement and inactivity costs.

While the inspection problem has been largely treated for single component systems, inspection policies become much more complex when considering multi-component systems. A two-component series system is considered in this paper.

The purpose of this paper is to formulate a new problem of the optimal design of a series manufacturing production line system, and to develop an efficient heuristic approach to solve it. The optimal design objective is to maximize the efficiency subject to a total cost constraint.

To estimate series production line efficiency, an analytical decomposition‐type approximation is used. The optimal design problem is formulated as one of combinatorial optimization where the decision variables are buffers and types of machines. This problem is solved by developing and demonstrating a problem‐specific ant system algorithm. Numerical examples illustrate the effectiveness of the algorithm.

It has been found that this algorithm can always find near‐optimal or optimal solutions quickly. The approach developed in this paper for manufacturing lines can be adapted for power systems and telecommunication systems.

The paper presents a new approach for the optimal design of buffered series production lines. This optimization approach aims at selecting both the machines and the levels of buffers. The paper also develops an efficient solution approach based on the ant system meta‐heuristic.

Reconfiguration mechanisms lead to the design of robust manufacturing systems which have the capability to allow the service continuity, in the presence of a failure, on the basis of a minimal degradation of performances. In this paper, a stochastic model is proposed to assess and to analyze the availability of reconfigurable systems whose equipments are subject to random failures. To distinguish between the normal behavior and the degraded one, the production rate is used as a performance measure. An availability model that takes into account the performance degradation is developed. Close form solutions of the steady‐state availability and the production rate of a reconfigurable system are calculated. Two optimization problems dealing with reconfigurable systems are also addressed. The paper considers a series system consisting of N stochastically independent components. Different technologies are assumed to be available for each component. The following design problems are studied: find the configuration, which allows maximizing the production rate of the system under resource constraints; and find the configuration that allows to reach some predetermined level of production rate at minimal cost. The design model of the first problem leads to mixed linear programming, while the design model of the second problem leads to integer linear programming. A numerical procedure is developed to solve both problems.

The authors consider a system which is a part of a complex equipment (e.g. aircraft, automobile, medical equipment, production machine, etc.), and which consists of N independent series subsystems. The purpose of this paper is to determine simultaneously the system design (reliability) and its preventive maintenance (PM) replacements periodicity which minimize the total average cost per time unit over the equipment useful life, taking into account a minimum required reliability level between consecutive replacements.

The problem is tackled in the context of reliability-based design (RBD) considering at the same time the burn-in of components, the warranty commitment and the maintenance strategy to be adopted. A mathematical model is developed to express the total average cost per time unit to be minimized under a reliability constraint. The total average cost includes the cost of acquiring and assembling components, the burn-in of each component, preventive and corrective replacements performed during the warranty and post-warranty periods. A numerical procedure is proposed to solve the problem.

For any given set of input data including components reliability, their cost and the costs of their preventive and corrective replacements, the system design (reliability) and the periodicity of preventive replacement during the post-warranty period is obtained such as the system’s total average cost per time unit is minimized. The obtained results clearly indicate that a decrease in the number of PM actions to be performed during the post-warranty period increases the number of components to be added at each subsystem at the design stage.

Given that the objective function (cost rate function) to be minimized is non-linear and involves several integer variables, it has not been possible to derive the optimal solution. A numerical procedure based on a heuristic approach has been proposed to solve the problem finding a nearly optimal solution for a given set of input data.

This paper offers to manufacturers a comprehensive approach to look for the most economical combination of the reliability level to be given to their products at the design stage, on one hand, and the PM policy to be adopted, on the other hand, given the offered warranty and service for the products and reliability requirements during the life cycle.

While the RBD problem has been largely treated, most of the published works have focussed on the development or the improvement of solving techniques used to find the optimal configuration. In this paper the authors provide a more comprehensive approach that considers simultaneously RBD, the burn-in and warranty periods, along with the maintenance policy to be adopted. The authors also consider the context of products whose component failures cannot be rectified through repair actions. They can only be fixed by replacement.

The purpose of this paper is to propose a framework to identify all the feasible disassembly sequences for a multi‐component product and to find an optimal disassembly sequence, according to specific criteria such as cost, duration, profit, etc.

Taking into account topological and geometrical constraints of a product structure, an AND/OR disassembly graph is built. Each graph node represents a feasible subassembly. Two nodes i and j are connected by an arc (i, j), called a transition, if the subassembly j can be obtained from the subassembly i by removing one or several connectors. Constraint programming approach is used to generate the feasible subassemblies and related transitions.

If a cost z_ij is incurred to perform a transition (i, j), an optimal disassembly sequence can be generated for a given subassembly, using the shortest path algorithm or a linear programming model.

The proposed approach performs very well compared to other approaches published in the literature, even when applied to products requiring parallel disassembly and including a large number of parts.

This approach has been successfully applied to assess the wheelchair maintainability at the design stage and will be implemented in CAD systems. One other application, regarding the disassembly process and total revenue maximization for product recycling, is now under consideration.

Applying constraint programming to efficiently generate the set of the feasible subassemblies constitutes the main contribution in this paper. This process is the hardest step in the disassembly sequencing problem.

This paper aims to investigate the optimization of the replacement with minimal repair policy for a system which experiences a time horizon of random length. Under such policy system replacement occurs at multiples of some period while minimal repair is performed at system failure between two successive replacements.

The objective function is the expected total cost composed of minimal repairs and replacements costs. A simple and compact expression is derived for the expected total costs and conditions under which an optimal replacement period exits are given. For sake of illustration, a numerical example is provided.

The paper finds that by the recent great technological development, the life cycle of present products is seen to be reduced more and more. This has motivated the development of maintenance optimization models for systems which experience an exact finite time horizon.

To ensure the benefits from the improved technologies, the information concerning the technological change must be taken into account. Such information is based on technological forecasting and difficult to obtain and merely rely on uncertainties.

The purpose of this paper is to extend the optimal design problem of series manufacturing production lines to series‐parallel lines, where redundant machines and in‐process buffers are both included to achieve a greater production rate. The objective is to maximize production rate subject to a total cost constraint.

An analytical method is proposed to evaluate the production rate, and an ant colony approach is developed to solve the problem. To estimate series‐parallel production line performance, each component (i.e. each set of parallel machines) of the original production line is approximated as a single unreliable machine. To determine the steady state behaviour of the resulting non‐homogeneous production line, it is first transformed into an approximately equivalent homogeneous line. Then, the well‐known Dallery‐David‐Xie algorithm (DDX) is used to solve the decomposition equations of the resulting (homogenous) line. The optimal design problem is formulated as a combinatorial optimisation one where the decision variables are buffers and types of machines, as well as the number of redundant machines. The effectiveness of the ant colony system approach is illustrated through numerical examples.

Simulation results show that the analytical approximation used to estimate series‐parallel production lines is very accurate. It has been found also that ant colonies can be extended to deal with the series‐parallel extension to determine near‐optimal or optimal solutions in a reasonable amount of time.

The model and the solution approach developed can be applied for optimal design of several industrial systems such as manufacturing lines and power production systems.

The paper presents an approach for the optimal design problem of series‐parallel manufacturing production lines.

The purpose of this paper is to investigate age replacement policies for two-component parallel system with stochastic dependence. The stochastic dependence considered, is modeled by a one-sided domino effect. The failure of component 1 at instant t may induce the failure of component 2 at instant t+τ with probability p _1→2. The time delay τ is a random variable with known probability density function h _{p 1→2} (.). The system is considered in a failed state when both components are failed. The proposed replacement policies suggest to replace the system upon failure or at age T whichever occurs first.

In the first policy, costs and durations associated with maintenance activities are supposed to be constant. In the second replacement policy, the preventive replacement cost depends on the system’s state and age. The expected cost per unit of time over an infinite span is derived and numerical examples are presented.

In this paper and especially in the second policy, the authors find that the authors can get a more economical policy if the authors consider that the preventive replacement cost is not constant but depends on T.

In this paper, the authors take into account of the stochastic dependence between system components. This dependence affects the global reliability of the system and replacement’s periodicity. It can be used to measure the performance of the system et introduced into design phase of the system.