Component Reliability Estimates Research Articles

Correction to: Estimation of component reliability from superposed renewal processes by means of latent variables

Correction to: Estimation of component reliability from superposed renewal processes by means of latent variables

Estimation of component reliability from superposed renewal processes by means of latent variables

Estimation of component reliability from superposed renewal processes by means of latent variables

A system reliability estimation method by the fourth moment saddle point approximation and copula functions

AbstractDespite many advances in the field of computational system reliability analysis, estimating the joint probability distribution of correlated non‐normal state variables on the basis of incomplete statistical data brings great challenges for engineers. To avoid multidimensional integration, system reliability estimation usually requires the calculation of marginal failure probability and joint failure probability. The current article proposed an integrated approach for estimating system reliability on the basis of the high moment method, saddle point approximation, and copulas. First, the statistic moment estimation based on the stochastic perturbation theory is presented. Thereafter, by constructing CGF (concise cumulant generating function) for the state variable with its first four statistical moments, a fourth moment saddle point approximation method is established for the component reliability estimation. Second, the copula theory is briefly introduced and extensively utilized two‐dimensional copulas are presented. The best fit copula for estimating the probability of system failure is selected according to the AIC (Akaike Information Criterion). Finally, the derived method is applied to three numerical examples for the sake of a comprehensive validation.

A novel approach based on meta-modeling technique and time transformation function for reliability analysis of upgraded automotive components

Early reliability estimation is still a challenging task. The paper presents a novel approach to deal with early reliability estimation of upgraded automotive components. The key idea is to combine reliability analysis based on efficient surrogate models and time transformation function principle. The surrogate model, built using Dimensional Decomposition Method and projection throughout a Lagrange polynomial basis, is used to substitute a time consuming implicit model initially used to compute the fatigue lifetime. The time transformation function is represented by a parametric power law model where the corresponding parameters are obtained through statistical analysis based on both numerical and experimental reliability results of a reference design. The reliability of an upgraded design is easily obtained by applying the time transformation function to the reliability estimation given by performing Monte-Carlo simulations on the surrogate model corresponding to the upgraded design. An application to a mechanical component, used in car seats, clearly illustrates the efficiency and the accuracy of the proposed approach.

Component reliability estimation based on system failure-time data

ABSTRACT We consider a coherent system consists of n components with independent and identically distributed lifetimes. We use the well-known signature-based representation of the system reliability to estimate the reliability of components of the system when the data is the system failure times collected according to a progressively censored scheme. Different estimators for the reliability of components are proposed and their properties are examined. A Monte Carlo simulation study have been carried out to compare the performance of the estimation methods.

Game theory based solution selection for multi-objective redundancy allocation in interval-valued problem parameters

This paper presents a multi-objective redundancy allocation problem (MORAP) for maximizing system reliability and simultaneously minimizing system cost with limitations on system entropy, which is essential for achieving system stability and sustainability. Entropy is an important part of this new model because it provides a measure of randomness, which should be limited to avoid risky solutions. Both component reliability estimates and component cost estimates are considered to formulate the model in more realistic sense. In real-life MORAP, the Pareto optimal set can be extremely large. A selection procedure based on the notion of game theory is then proposed to determine representative solutions for solving the problem. For MORAP, the game theoretical framework can select representative solutions effectively with higher system reliability, lower associated variance of the reliability estimate and higher system entropy. The validity and the performance of the proposed approach are tested through three numerical examples, and computational results are analyzed.

Multiscale Modeling and Analysis of DC/DC Converter Based on Macro and Micro-Scale Description

Designing of power electronics converters seldom considers multiscale coupling relationship between components and circuits. Although macroscale is computationally efficient, it barely contains the micro implications about aging, deterioration, transient performance, etc. Conversely, the microscale obtains multiple observation scope, however, with computational disaster. In this paper, the multiscale method is applied in the power electronic modeling which provides a qualitative methodology to combine the scales appropriately, and to obtain the dynamic response under the action of multiphysics accurately. Firstly, the concept of multiscale and the scale division of power electronic system are expounded. Considered the dynamic behavior of components and converters with different complexity, the framework of multiscale converters and modeling method is proposed. Afterwards, a scalable coupling separation method is presented, providing interfaces for the multiphysics and dielectric physics. Additionally, a multiscale conversion algorithm considering both efficiency and accuracy is designed. In the end, a DC/DC converter is taken for the verification. The feasibility and advantages of multiscale method and conversion algorithm have been demonstrated in accuracy, consistency and multiple analysis. By filling in the microscale observation, it is plausible to provide theoretical guidance for reducing margin, accurate analysis and reliability estimation of electronic components.

Reliability Analysis for Load-Sharing Parallel Systems with No Failure of Components

Conventional reliability analysis of load-sharing parallel systems is mainly based on the assumption that there is failure of components. When a component fails, the load redistributes among the remaining surviving components. But sometimes there is no failure of components during the usage of a load-sharing parallel system. When there is no failure of components, working load is dispatched among components, which means the load on each component is almost the same during the whole working process. The degradation process of each component is also almost the same. The reliability analysis of a load-sharing parallel system with no component fails can be simplified to the reliability analysis of a component. In this paper, three cases are studied. First, based on stress-strength interference model, estimation of component reliability with load application times is generated, which considered the nonlinear fatigue damage model. Then, Poisson process is used to describe reliability changed with time. Third, the reliability is studied when the working load is not constant but changed with time or load application times. Some examples are used to illustrate the application of these models, and the Monte Carlo simulation method is used as standard to verify the proposed models.

Integrating renewal process modeling with Probabilistic Physics-of-Failure: Application to Loss of Coolant Accident (LOCA) frequency estimations in nuclear power plants

Renewal process modeling is used for the failure prediction of hardware components under periodic maintenance. While most studies utilized data-driven approaches to estimate the input parameters for renewal process models, this paper initiates a line of research to integrate renewal process modeling with probabilistic models of underlying mechanisms associated with physical degradation and maintenance. At this stage of the research, the methodology integrates Markov modeling with Probabilistic Physics-of-Failure (PPoF) models of degradation, while maintenance is treated by a data-driven approach. This methodology is valuable to obtain a more accurate estimation of component reliability and availablity, especially when (i) components are highly reliable, and failure data are limited, (ii) historical data are unreliable due to changes in design, operation, and maintenance, or (iii) advanced technologies have emerged limiting operational data. The methodology explicitly incorporates the underlying spatiotemporal causes of failure into the renewal model, allowing to rank the criticality of causal factors to improve maintenance and mitigation strategies. Although the new methodology is applicable for component reliability and availability analysis in diverse industries, this paper demonstrates its value for estimating frequencies of a Loss-Of-Coolant Accident (LOCA), which is an initiating event in Probabilistic Risk Assessment (PRA) of Nuclear Power Plants (NPPs).

Estimation of Component Reliability in Coherent Systems With Masked Data

The reliability of a coherent system of components depends on the reliability of each component and the initial statistical work should be an estimation of the reliability of each component. This paper represents a challenging task because if the system fails, the failure time of a given component cannot be observed, that is, the phenomenon of censored data occurs. A solution for the reliability estimation of components exists when the system failure time and the status of each component are available at the time of system failure. However, it may be difficult to identify the status of the components at the moment of system failure. Such cases represent systems with masked causes of failure. Since parallel and series systems are the simplest systems, numerous solutions have been reported in the literature. To the best of our knowledge, this paper is the first to present the general case of coherent systems without the restriction of an identically distributed lifetime. The three-parameter Weibull Bayesian model is proposed. The Gibbs with the Metropolis-Hasting algorithm supports the statistical work of obtaining the posterior distribution quantities. With several simulations, the excellent performance of the model is evaluated. A real dataset of computer hard drives is analyzed to show the practical relevance of the proposed model.

Mining Spare Parts Provision by Reliability Analysis, Case Study: Sungun Copper Mine

Summary Maintenance and spare parts support are two basic and critical issues of support to product, because often due to lack of technology and other compelling factors (like economy, environmental situations, etc), in the design phase it is impossible to design a product that will fulfill its function. Therefore, the need for support haw become vital to enhance system effectiveness and prevent unexpected failure. This paper was focused on second issues of support. Thus, for studying of the required spare parts, we analyzed the failure time data for a period of 11 years, which has been obtained from Sungun Copper mine in Iran. The expected number of the required spare tires in one year (three working shifts per day) when , and hours for lower, optimum and upper value are considered to be the real mean time to failure of the tire with a 95 percent confidence of availability are equal to 3.2, 2 and 1.2 respectively. Then, the economic order quantity with respect to annual demand rate is calculated. Introduction Product support falls into two broad categories, namely support to the customer and support to the product. The research presented in this paper is focused on support to the product, which is greatly influenced by the product reliability characteristics. Spare parts organize one of the product support issues that can be divided into two types, namely repairable and non-repairable. In the present research we deal with non-repairable parts. Spare parts provisioning is a complex problem and requires an accurate analysis of all factors that may affect the required number of spare parts. Different reliability based statistical approaches have been developed for spare parts provision. The reliability-based spare parts provision is the most common one. Methodology and Approaches The methodology is based on three main tasks, 1) Definition of boundaries, assumptions, and data collection.2) Identification of analysis approach and estimation of component reliability and failure rate characteristics. 3) Required spare parts calculation and inventory management requirements Results and Conclusions In this paper, a methodology is proposed to consider reliability in the required spare parts tires estimation. The result of analysis for optimum estimation of PLP parameters presented that for a year dump truck need 2 new tires. For inventory management whenever the inventory position reaches 1 units/dump truck, we should order 2 units/dump truck.

Bayesian Estimation of Component Reliability in Coherent Systems

The first step in statistical reliability studies of coherent systems is the estimation of the reliability of each system component. For the cases of parallel and series systems, the literature is abundant, but it seems that the present paper is the first to present the general case of component inferences in coherent systems. The failure time model considered here is the three-parameter Weibull distribution. Furthermore, identically distributed failure times are not a required restriction. An important result is proved: without the assumption that components' lifetimes are mutually independent, a given set of sub-reliability functions does not identify the corresponding marginal reliability function. The proposed model is general in the sense that it can be used for components of any coherent system, from the simplest to the most complex designs. It can be considered for all kinds of censored data, including interval-censored data. An important property obtained for the Weibull model is that the posterior distributions are proper, even for non-informative priors. Using several simulations, the excellent performance of the model is illustrated. As real examples, boys' first use of marijuana and a device from a field-tracking data set are considered to show the efficiency of the solution even when censored data occurs.

Estimating Component Reliability Based on Failure Time Data from a System of Unknown Design

Estimating Component Reliability Based on Failure Time Data from a System of Unknown Design

Nonparametric estimation of component reliability based on lifetime data from systems of varying design

Nonparametric estimation of component reliability based on lifetime data from systems of varying design

Nonparametric estimation of component reliability based on lifetime data from systems of varying design

Failure time data collected from fielded systems provide indirect information about the performance of the system’s components. Since it is often difficult to create or simulate field conditions in laboratory settings, the process of drawing inferences about component characteristics from data on system performance is of practical importance. However, there is very little literature on this problem that treats such inferences from a nonparametric perspective, and less literature still that allows the systems of interest to be of arbitrary design. The present paper focuses on nonparametric estimation of a common component reliability function using independent samples from coherent systems of varying design whose components have independent, identically distributed lifetimes. Two estimation approaches are studied. The first is conventional, and is based on treating each of the estimation problems separately; it is shown that these mixture estimators are consistent, and their asymptotic behaviours are characterized. The second estimator is quite unconventional. It is obtained by solving multiple point-wise maximum likelihood estimation problems simultaneously, and combining the separate estimators, each at fixed time points, to obtain an overall estimator of the reliability function. We show that the latter approach produces a legitimate reliability function and that, asymptotically, it is uniformly superior to all the estimators of the first type. Related estimators of the lifetime density and failure rate functions are also obtained, and their theoretical and numerical properties are described.

Nonparametric estimation of component reliability based on lifetime data from systems of varying design

Nonparametric estimation of component reliability based on lifetime data from systems of varying design

Weighted Estimation of Component Reliability in Series Systems

The characteristics of the failure times of components in a series system are estimated from observed system failure times and causes of failure. Usually this situation is also approached from a competing risks model viewpoint where, in case of statistical independence, the lifetime of each component may be estimated by the Kaplan-Meier method (KM) . For simplicity, we consider only two components in the system. If we are interested in estimating the reliability or survivor function of each component, we may construct for each case an estimator by conveniently interpreting the data set. In this paper, we suggest a modified version of the KM method for estimating the survivor function of a component that uses the information about the lifetime of the other component (which is not being analysed at the moment) to get a better accuracy with heavily censored sampling information, which in fact is the case of the reliability context addressed here.

Effects of Uncertainty in Both Component Reliability and Load Demand on Multistate System Reliability

Considerable studies have been done on system reliability estimation under uncertainty. Most of these studies focus on binary-state systems, and the uncertainty source is only admitted in the component reliability estimation. The studies on multistate systems (MSSs) do exist, but these studies either completely ignore the uncertainty in load demand during the system operation period or are restricted to parallel system structures. This paper attempts to extend these previous studies and investigate the effects of uncertainty in both component reliability and load demand on system reliability for general MSSs with any series-parallel structures. Analytical expressions for calculating the expected value and the variance of system reliability considering these two sets of uncertainty factors are derived based on the universal generating function method in an iterative form. The accuracy of the proposed analytical approach is validated through Monte Carlo simulations. Various examples on a complex series-parallel MSS are used to investigate the effects of the two sets of uncertainty factors. The results indicate that these two sets of uncertainty factors affect the expected value and the variance of system reliability for MSSs in different ways.

Predictive Analysis by Incorporating Uncertainty through a Family of Models Calibrated with Structural Health-Monitoring Data

AbstractComplex analysis and design of structures, especially landmark structures such as long-span bridges, have been conducted by many engineers and researchers. Currently, it is possible to collect more and precise monitoring data as well as to develop complex three-dimensional (3D) FEM models. These models, which can be calibrated using structural health-monitoring (SHM) data, can be used for the estimation of component and system reliability of bridges. However, the uncertainties related to the data, analysis, and nonstationary nature of the structural behavior need to be better incorporated by using a set of models that are continuously updated with monitoring data. This set of models constitutes a family as a result of the approach by which the models are obtained and the relationships among them. The objective of this paper is to explore the impact of uncertainty in predicting the system reliability obtained by a one-time, initially calibrated FEM model as well as by a family of FEM models continu...

Designing Reliability Demonstration Tests for One-Shot Systems Under Zero Component Failures

It is difficult to estimate the confidence interval of a system's reliability with zero failures experienced. We approach this problem by proposing a hybrid model that integrates the Bayesian model with the variance propagation technique. The Bayesian model will compute the moments of component reliability estimates, and the variance propagation technique is used to estimate the system reliability variance. The confidence interval for the system reliability is then derived by matching the moments with a beta distribution. As a major contribution, the distribution for reliability estimates with zero failures is explicitly derived. The performance of the new model is compared with existing methods, and further validated by simulation data. The results show that the hybrid model generally outperforms existing methods in terms of estimation accuracy. Because the new model does not require multiple integral calculations, it can be applied to design complex systems configured in mixed series-parallel or networked components.