Proceedings of the Institution of Mechanical Engineers,Part O: Journal of Risk and Reliability

This article proposes a novel preventive replacement policy based on condition monitoring and imperfect manual inspection for systems subject to a two-stage deterioration process, where the two-stage deterioration process is modeled by the white noise process and Brownian motion with a drift, respectively. The proposed preventive replacement policy is implemented using two thresholds: a failure threshold and a preventive threshold. Specifically, if the condition monitoring measurement is observed to cross the failure threshold, then the failure replacement will be carried out; and, if the condition monitoring measurement is observed to cross the preventive threshold while lower than the failure threshold, then the system needs to be checked by manual inspection, and the preventive replacement will be carried out once the system is found to be in the defective state. In this article, we consider that manual inspection is imperfect, namely, there is a probability that the defect will be unnoticed. By minimizing the expected cost per unit time, we obtain the optimal condition monitoring interval and preventive threshold. A numerical example is provided to demonstrate the performance of the proposed condition-based replacement policy. Comparisons are made with the existing work, which shows the effectiveness and superiority of the proposed policy.

Due to the limitations of space or extra costs incurred, the reliability optimization problem of a spare system is of great interest and importance. In this article, we devote our efforts into the investigation of reliability optimization problem of the warm spare gate and cold spare gate. For a spare gate with fixed components, we first examine the relationship between the component order and the corresponding reliability; then, the equivalence of a cascaded model with a multiple-input spare gate is further presented. We find that for a warm spare gate, the corresponding reliability value is anticipated to be affected by the adopted component replacing order; nevertheless, the reliability is a fixed value once the components are provided for a cold spare gate. This finding indicates that reliability is irrespective of the component order for a cold spare gate. Therefore, for the warm spare gate, the component order can be varied to improve the corresponding system reliability, whereas for the cold spare gate, we should attempt to improve the reliabilities of the spare components aiming to obtain a higher reliability. These findings are potentially useful in the design process of a system consisting of spare gates.

This article deals with the estimation of parameters for the load-sharing parallel system when the underlying distribution of component lifetime is exponential Pareto. The appropriate iterative method is used to get the parameter estimators. And the asymptotic distribution of the proposed maximum likelihood estimator is presented. For the model simplified, the maximum likelihood and best asymptotic unbiased estimators of parameters are obtained. Moreover, the confidence intervals for the parameters are constructed using the parameter bootstrap technique. In the end, some appropriate simulation methods and an example are used to substantiate the proposed methods.

Due to that the operating environment is becoming more and more complex and rigorous, the multiple competing failure modes for the reliability system is much commonly seen. In order to improve the system performance, a sensor-based degradation calibration policy (SBDC policy) is presented in this paper. The model considers the competing failure process which is described by the soft and hard failure modes. In detail, the soft failures occur when the degradation of the system exceeds the failure threshold, and the hard failures are caused by the same shock process. We use the Wiener process model to describe the soft failure and the shock process to describe the catastrophic failure. Meanwhile, in the shock process, the damage associated with the system is normal distributed which is related to the duration of the adjacent shocks. This extended model with calibrations has a good application value for the corresponding complex reliability systems which are subject to the dependent competing failure modes. By the model in this article, the system reliability and safety can be improved and the risk of the abrupt damage shall be reduced as the circumstance changes.

In theory and practice, system performance is one of the most important issues. Therefore, a series of indexes has been proposed for evaluating the system performance, such as availability. However, these indexes still cannot meet the variant requirements in the reliability and other fields. The purpose of the article is to develop some theoretical results that may be used in modeling the evolution of system performance. So, based on the aggregated stochastic process theory, some new indexes are introduced and established in Markov repairable systems. In this model, the state space is partitioned into working subset W and failure subset F. The system is regarded as stable if the state of system enters one subset, either W or F, at any instance and sojourns within the subset exceeding a given non-negative threshold τ . Otherwise, the system is regarded as unstable. Under these assumptions, the concepts of point-wise and interval-wise are proposed, and the computation formulae of two types of indexes are derived in the theory. Finally, a special case and a few of numerical examples are given to illustrate the results obtained in the paper.

Reconfigurable systems can meet the changing requirements of system performance by several approaches, such as adjusting the system structure, improving the component performance, and reassigning components. However, it is also challengeable to find a cost-effective maintenance scheme by integrating these maintenance approaches. This article investigates the multi-objective maintenance optimization problem for reconfigurable systems with the consideration of maintenance cost and system reliability. First, the multi-objective maintenance optimization model is established to maximize the system reliability and minimize the total maintenance cost considering the constraints on budget and system performance. Second, a multi-objective Birnbaum importance is proposed to quantify the contribution of the individual component to the system reliability. The multi-objective Birnbaum importance–based non-dominated sorting genetic algorithm II is developed to obtain the optimal maintenance scheme with the maximum system reliability and minimum maintenance cost. Finally, the performance of multi-objective Birnbaum importance–based non-dominated sorting genetic algorithm II is proved by three numerical experiments. Experiment 1 verifies the advantage of multi-objective Birnbaum importance compared with Birnbaum importance to improve the system reliability in direct maintenance. Experiment 2 shows that the effectiveness of multi-objective Birnbaum importance is much better than that of the Birnbaum importance to enhance the performance of non-dominated sorting genetic algorithm II in comprehensive maintenance. Experiment 3 illustrates that the performance of multi-objective Birnbaum importance–based non-dominated sorting genetic algorithm II is better than that of other multi-objective algorithms combining with multi-objective Birnbaum importance.

In this article, the reliability of network systems subject to probabilistic propagation failure and failure isolation effects is considered. Probabilistic propagation failure is the failure of some components in a system, which will cause other components to fail with certain probabilities. Probabilistic propagation failure exists in various network systems, such as computing network system and nuclear power generating network system. Failure isolation means that the failure of a trigger component will lead to its corresponding dependent components being isolated from the network system. Since the failure isolation effect is activated only when the failure of trigger components occurs before the occurrence of probabilistic propagation failure, there exists a competing failure in the time domain between the failure of a trigger component and the components with probabilistic propagation failure effect. If a trigger component failure occurs first, the system is insensitive to any failures of components being isolated. In this article, a combinatorial method based on binary decision diagram is proposed to analyze the reliability of the network systems subject to probabilistic propagation failure and failure isolation effects. The method can be applied to any network system and any type of lifetime distribution of the system components. As an example, a wide area network system is analyzed. Some numerical results about reliability indexes are provided to verify the feasibility and accuracy of the proposed method.

It is desired to build the life distribution models of critical components (which are assumed to be non-repairable) of a repairable system as early as possible based on field failure data in order to optimize the operation and maintenance decisions of the components. When the number of the systems under observation is large and the observation duration is relatively short, the samples obtained for modeling are large and heavily censored. For such samples, the classical parameter estimation methods (e.g. maximum likelihood method and least square method) do not provide robust estimates. To address this issue, this article develops a hybrid censoring index to quantitatively describe censoring characteristics of a data set, proposes a novel parameter estimation method based on information extracted from censored observations, and evaluates the accuracy and robustness of the proposed method through a numerical experiment. Its applicable range in terms of the hybrid censoring index is determined through an accuracy analysis. The experiment results show that the proposed approach provides much accurate estimates than the classical methods for heavily censored data. A real-world example is also included.

With the introduction of reliability engineering, electrical power system reliability has become an important basis for decision-making in the power industry. Two operation cases of electrical power systems are considered in this article. When the system is in an ordinary way, the influence between two system components will affect the importance measure of one component. When some component is in maintenance, preventive maintenance for working components and corrective maintenance for failed components can be executed simultaneously to enhance electrical power system performance. In view of the above two cases, two importance measures are proposed to effectively guide the preventive maintenance, aiming to improve the system performance within a limited budget. Reliability analysis procedure and methods applied toward the two importance measures are then developed and illustrated with the analysis on a Dual Element Spot Network system with double power supplies and double loads. Finally, a strategy for preventive maintenance is proposed by ranking the importance of these components.

This article focuses on a toroidal connected-(r,s)-out-of-(m,n):F lattice system, which has m vertical circles and n horizontal circles, and the components are deployed at their intersections. This system fails if and only if the system has an ( r , s ) sub-matrix where all components fail. It might be used to evaluate the reliability of a system with a toroidal structure, for example, toroidal storage tanks and neutron accelerators. The purpose of this article is to efficiently compute the reliability of the toroidal connected-(r,s)-out-of-(m,n):F lattice system. First, we propose a recursive method and then develop two kinds of algorithms based on this method. We apply an idea of preparing a set in a pre-processing phase to one algorithm, and consequently, the algorithm with the idea requires the extra memory space but has better time complexity compared with the other one. Finally, we investigate the efficiency of the two algorithms through numerical experiments.

We deal with an optimization problem of planning a maintenance strategy for 145 kV gas-insulated switchgear and apply reliability-centered maintenance approach to find an efficient and effective strategy. In a three-step sequential process, we use failure mode effects and criticality analysis to define the critical failure modes that cause the major failures of 145 kV gas-insulated switchgear, select the maintenance significant items to remove the major failure modes, and finally apply logic tree analysis to assign the appropriate maintenance task to each critical failure modes in the system. We then compare the assigned maintenance tasks with existing tasks in the installation, operation, and maintenance manuals developed in the system design and development phase. The assigned maintenance tasks in this study are compared with those in the system design and development phase by simulation, and a simple heuristic method is proposed to find optimal solutions.

Renewal-geometric process is used to describe such a non-homogeneous deteriorating process that a system will deteriorate after several consecutive repairs, not after each repair described by the geometric process. In the maintenance domain, the effect of corrective maintenance after failure is generally not repairable as new (e.g. geometrically deteriorating). Preventive maintenance is critical before a system failure, due to economic losses and security threats caused by a sudden shutdown. Therefore, this article assumes that a system is geometrically deteriorating after corrective maintenance, wherein preventive maintenances sequence in the same repair period form a renewal process since it can restore the system to the initial state of the period. Furthermore, a binary policy ( N , T ) is utilized to minimize the long-run average cost rate, where N represents the number corrective maintenances and T denotes the time interval between two consecutive preventive maintenances. In particular, pseudo-age replacement model represents a special case of N = 1 , which is considered as a generalization of the traditional age-based replacement model. Subsequently, the optimal policy N * can be verified in theory and an asymptotic optimal policy ( N * , T * ) can be obtained based on a heuristic grid search. Finally, numerical examples verify the effectiveness of this proposed model and show that implementation of preventive maintenance for some repairable systems is superior to no preventive maintenance in both economic and reliability aspects.

Verification and validation plays a crucial role in the new product development process since it is able to assure product performance and eventually determine customer satisfaction. The verification and validation planning assigns a set of verification and validation activities such as computational simulations and physical tests, which are proposed to mitigate the design risks of specific failure modes, to verify and validate that the product conforms to the design objectives. This article formulates the optimal verification and validation planning using set covering, set partition, and set packing concepts. An extended optimization model based on set covering is also formulated, which can well accommodate the requirements including minimizing the overall risk of the product design, meeting pre-specified risk thresholds of specific failure modes, and covering identified critical failure modes. Additional constraints such as the implementation sequences, time gaps of various types of verification and validation activities, and the distinct effectiveness of each verification and validation activity in reducing design risks are considered in the verification and validation planning optimization models. The decay of the improvement effectiveness of a failure mode with multiple verification and validation activities over time is also considered. The application of the proposed mathematical optimization models for product verification and validation planning is illustrated through the product development of a power generation system within a diesel engine.