Limited Failure Data Research Articles

Grey Duane model for reliability growth prediction under small sample uncertain failure data

Traditional reliability growth models require failure data to adhere to a specific distribution, which greatly limits the applicability in data-scarce scenarios. Drawing from the characteristic in modeling small sample and poor information of grey system theory, this study introduces a novel grey Duane reliability growth prediction model, tailored for analyzing uncertainty in limited failure data. The characteristics and adaptability of the proposed grey Duane model (GDM) are analyzed and compared with two traditional reliability growth prediction models. Utilizing a first-order accumulation generation operation, a differential function is established to estimate the unknown parameters in GDM. The proposed GDM enables synchronized predictions of failure time, failure number, and instantaneous mean time between failures. To validate the applicability of GDM in reliability growth management, an industrial case in particular electronic equipment during an aircraft’s flight-testing process is applied, whose results demonstrate its superior predictive accuracy compared to alternative models.

Bootstrap-Based Resampling Methods for Software Reliability Measurement Under Small Sample Condition

Highly reliable software systems rarely fail during tests because they are usually designed with fault-tolerant mechanisms and tested comprehensively. It is usually difficult to obtain sufficient failure data to carry out reliability measurements by using traditional software reliability models. These models are typically based on probabilistic statistics, and the measurement accuracy cannot be guaranteed with insufficient failure data. We propose a nonparametric bootstrap (NBP) resampling method and six parametric bootstrap (PB) resampling methods to construct software reliability models for small sample conditions based on commonly used models, i.e., the Jelinski–Moranda (J–M), Goel–Okumoto (G–O), Musa–Okumoto (M–O), Schneidewind, Duane and Littlewood-Verrall models. The bootstrap is a statistical procedure that resamples a single dataset to create many simulated samples. Our experimental results on fourteen failure datasets collected from industry and academia show that the proposed models improve by 10.2–18.0% failure time prediction accuracy, 24.7–30.7% curve fitting accuracy, and 7.7–42.9% reliability measurement accuracy compared with the original models. Furthermore, our approaches achieve 58.3–91.1% better failure time prediction accuracy in the case of small sample conditions compared to state-of-the-art machine learning and neural network-based methods. Overall, our approaches can perform more accurate reliability measurements than the original models even in scenarios with limited failure data.

Reliability Allocation in Complex Systems Using Fuzzy Logic Modules

Reliability allocation is critical in the design of complex engineering systems. However, traditional allocation methods rely on precise failure data and have difficulty handling epistemic uncertainty. This paper proposes using fuzzy logic modules for reliability allocation in complex systems. Fuzzy logic can represent imprecise data using membership functions and incorporate expert knowledge through fuzzy rules. The methodology involves developing fuzzy logic modules for each subsystem to allocate reliability based on fuzzy input variables like failure rate and criticality. The outputs are aggregated to obtain the system level allocation. A case study demonstrates the approach and compares results against traditional methods. The fuzzy logic modules are shown to optimize reliability allocation under epistemic uncertainty. This paper demonstrates the advantages of using fuzzy logic for reliability allocation in complex systems with limited failure data. The methodology provides a new tool for reliability engineers to handle imprecise information and optimize designs.

Limited Fault Data Augmentation With Compressed Sensing for Bearing Fault Diagnosis

Sufficient data is necessary for intelligent fault diagnostic approaches. However, in practice, it is often the case that only limited fault data is available due to various reasons, making it a challenge to accurately identify the health condition of bearings. To deal with the limited fault data issue, data augmentation strategies, such as generative adversarial network (GAN), are widely utilized. However, GAN has the disadvantages of being difficult to train and restricted ability to generate new data when the fault sample size is limited. Specifically, GAN requires a long training time and abundant training data to make the distribution of generated data closer to the distribution of actual data. This article presents a novel data augmentation approach with compressed sensing for fault diagnosis of bearings to better address the issue of limited fault data. The generated data by compressed sensing is diverse. In addition, the generated data is highly similar to the original data in frequency domain, thus retaining the main feature information of the original data. Furthermore, data augmentation achieved through compressed sensing requires less training data and has lower computational complexity. For bearing fault diagnosis under limited failure data, the limited fault data is first augmented based on compressed sensing, allowing for high fidelity reconstruction and high diversity data generation. Then, the augmented data is utilized to train a deep convolutional neural network to automatically learn and extract features for fault identification. The effectiveness of the presented approach is verified using two bearing datasets.

멀티 세그멘테이션과 태깅을 이용한 딥러닝 기반 송변전설비 학습 방법 연구

Recently, with the development of deep learning technology, failure analysis and failure diagnosis research using image analysis of objects have been actively conducted. In particular, research on algorithm and system development for diagnosing facilities using drone photographed images is being applied to the industrial field. The results are reaching the level of commercialization. In the electricity field, drone images have been used in the field of power facility diagnosis since two to three years ago. There are not many abnormal learning data to determine whether transmission facilities are abnormal, so full-scale use in the actual field is limited. Therefore, this study proposes a method of securing more learning data by utilizing images of limited failure data. In addition, the obtained data is used for learning to present deep learning methods and research results for developing a more accurate transmission facility diagnosis system. As a result of this study, it was confirmed that the average precision was improved by about twice from 39.2% to 81.1% by applying the learning model technology. This improved method of artificial intelligence learning technology is expected to prevent power transmission failure in advance, avoid power outage costs caused by failure, and reduce maintenance costs through inspection automation.

Failure assessment of corrosion affected pipeline networks with limited failure data availability

Many pipeline networks around the world do not have a failure history. Because of financial reasons or lack of technology, this is a common issue for asset managers in developing and under-developed countries. In the case of data scarcity, it is not straightforward to assess the risk and reliability of such systems. This research proposes a probabilistic framework for analysing the reliability of corroded pipeline networks with limited failure history availability. To achieve this aim through a comprehensive reliability assessment methodology, state-of-the-art corrosion modeling, component and system failure analysis techniques are utilized. A variety of spatio-temporal reliability assessment methods are employed along with the random-field representation of spatially variable parameters to estimate the failure in a large-scale corrosion affected pipeline network. The applicability of the methodology is evaluated through a case-study of buried pipeline network, and the results are verified with failure history data reported by the industry. The results obtained from the proposed methodology, combined with the cost data provided by asset managers make accurate risk assessment of process/water pipeline possible.

An Adaptive Bayesian Melding Method for Reliability Evaluation Via Limited Failure Data: An Application to the Servo Turret

In the early stage of product development, reliability evaluation is an indispensable step before launching a product onto the market. It is not realistic to evaluate the reliability of a new product by a host of reliability tests due to the limiting factors of time and test costs. Evaluating the reliability of products in a short time is a challenging problem. In this paper, an approach is proposed that combines a group of experts’ judgments and limited failure data. Novel features of this approach are that it can reflect various kinds of information without considering the individual weight and reduces aggregation error in the uncertainty quantification of multiple inconsistent pieces of information. First, an expert system is established by the Bayesian best–worst method and fuzzy logic inference, which collects and aggregates a group of expert opinions to estimate the reliability improvement factor. Then, an adaptive Bayesian melding method is investigated to generate a posterior by inaccurate prior knowledge and limited test data; this method is made more computationally efficient by implementing an improved sampling importance resampling algorithm. Finally, an application for the reliability evaluation of a subsystem of a CNC lathe is discussed to illustrate the framework, which is shown to validate the reasonability and robustness of our proposal.

Using Expert Knowledge to Generate Data for Broadband Line Prognostics Under Limited Failure Data Availability

Due to exposure to the driving rain, water ingress can cause faults in electrical joints, junctions and distribution points in broadband lines. Over time, faulting behaviour may grow in magnitude eroding the electrical capability of these lines causing degradation of broadband service. Developing effective data-driven models for broadband line prognostics remains a challenge due to the limited failure data availability in the telecommunications industry. In order to address this problem, we present a technique for generating failure data that realistically reflect the behaviour of degrading broadband lines. To this end, we use the conditional generative adversarial network and more importantly, we control and direct the failure data generation process using expert knowledge on the water ingress failure cause. The proposed technique is evaluated using a real-world case study involving the time-to-failure prediction of two types of broadband lines in a south-west city in England. The prognostics performance is measured using the Kappa statistic and F-score. Benchmark performance is obtained using Random Oversampling, Synthetic Minority Oversampling and Adaptive Synthesis which can be used to oversample failure data by duplicating existing failure data or randomly generating data. Among these techniques, Random Oversampling achieved the best prognostics performance. It is shown that the proposed technique outperforms Random Oversampling technique by a large margin. More specifically, it increased the prognostics performance by 33% (Kappa statistic) and 25% (F-score) for Asymmetric Digital Subscriber Lines, and 17% (Kappa statistic) and 13% (F-score) for Very High Bitrate Digital Subscriber Lines compared to the Random Oversampling technique.

A Methodology for Prognostics Under the Conditions of Limited Failure Data Availability

When failure data are limited, data-driven prognostics solutions underperform since the number of failure data samples is insufficient for training prognostics models effectively. In order to address this problem, we present a novel methodology for generating failure data which allows training datasets to be augmented so that the number of failure data samples is increased. In contrast to existing data generation techniques which duplicate or randomly generate data, the proposed methodology is capable of generating new and realistic failure data samples. The methodology utilises the conditional generative adversarial network and auxiliary information pertaining to failure modes to control and direct the failure data generation process. The theoretical foundation of the methodology in a non-parametric setting is presented and we show that it holds in practice using empirical results. The methodology is evaluated in a real-world case study involving the prediction of air purge valve failures in heavy-trucks. Two prognostics models are developed using the gradient boosting machine and random forest classifiers. When these models are trained on the augmented training dataset, they outperformed the best solution previously proposed in the literature for the case study by a large margin. More specifically, costs due to breakdowns and false alarms are reduced by 44%.

Small sample reliability growth modeling using a grey systems model

ABSTRACTWhen performing system-level developmental testing, time and expenses generally warrant a small sample size for failure data. Upon failure discovery, redesigns and/or corrective actions can be implemented to improve system reliability. Current methods for estimating reliability growth, namely the Crow (AMSAA) growth model, stipulate that parameter estimates have a great level of uncertainty when dealing with small sample sizes. For purposes of handling limited failure data, we propose the use of a modified GM(1,1) model to predict system reliability growth parameters and investigate how parameter estimates are affected by systems whose failures follow a poly-Weibull distribution. Monte-Carlo simulation is used to map the response surface of system reliability, and results are used to compare the accuracy of the modified GM(1,1) model to that of the AMSAA growth model. It is shown that with small sample sizes and multiple failure modes, the modified GM(1,1) model is more accurate than the AMSAA model for prediction of growth model parameters.

Helicopter blade reliability: Statistical data analysis and modeling

The concept of reliability has been attracting attentions in mechanical engineering following the developments of the aerospace industries. Limited failure data and statistical analyses of helicopter components reliability exist in the technical literature. For filling this gap, a nonparametric analysis is conducted on the performance of the 338 blades of some Iranian helicopters, which were in service between 1974 and 2012. These blades have 41 different failure modes. In this paper, statistical reliability analysis is conducted based on two strategies: In strategy I, general failure is defined as scrapping or retirement of the blade. In strategy II, the blade is assumed to be subjected to different modes of failure and the cumulative mode-specific functions are derived for each failure modes using Nelson–Aalen estimator. The Kaplan–Meier estimator is used for calculating the nonparametric reliability functions. Confidence intervals are derived for the reliability results and parametric fits are conducted using the maximum likelihood estimation. An important result from parametric analysis is that the blade reliability has a 3-parameter Weibull distribution and so the blades exhibit an increasing failure rate. Finally, considering the mode-specific hazard functions, the failure mode 1, i.e., excessive vibration is observed to have major contribution to the blade failures.

Design for Reliability of Complex System: Case Study of Horizontal Drilling Equipment with Limited Failure Data

Reliability is an important phase in durable system designs, specifically in the early phase of the product development. In this paper, a new methodology is proposed for complex systems’ design for reliability. Specific test and field failure data scarcity is evaluated here as a challenge to implement design for reliability of a new product. In the developed approach, modeling and simulation of the system are accomplished by using reliability block diagram (RBD) method. The generic data are corrected to account for the design and environment effects on the application. The integral methodology evaluates reliability of the system and assesses the importance of each component. In addition, the availability of the system was evaluated using Monte Carlo simulation. Available design alternatives with different components are analyzed for reliability optimization. Evaluating reliability of complex systems in competitive design attempts is one of the applications of this method. The advantage of this method is that it is applicable in early design phase where there is only limited failure data available. As a case study, horizontal drilling equipment is used for assessment of the proposed method. Benchmarking of the results with a system with more available failure and maintenance data verifies the effectiveness and performance quality of presented method.

Bridging the Gap between Quantitative and Qualitative Accelerated Life Tests

Test planners have long sought the ability to incorporate the results of highly accelerated life testing (HALT) into an early estimate of system reliability. While case studies attest to the effectiveness of HALT in producing reliable products, the capability to translate the test's limited failure data into a meaningful measure of reliability improvement remains elusive. Further, a review of quality and reliability literature indicates that confusion exists over what defines a HALT and how HALT differs from quantitative accelerated life testing methods. Despite many authors making a clear distinction between qualitative and quantitative accelerated life tests, an explanation as to why this delineation exists cannot be found. In this paper, we consider an exemplary HALT composed of a single stressor to show that the HALT philosophy precludes the estimation of a system's hazard rate function parameters because of the test's fix implementation strategy. Four common accelerated failure data analysis methods are highlighted to show their limitations with respect to estimating reliability from HALT data. Finally, a modified accelerated reliability growth test is proposed as a way forward for future research in HALT scenarios to characterize the risk of attaining a reliability requirement and improve parameter estimation. Copyright © 2014 John Wiley & Sons, Ltd.

A physical probabilistic model to predict failure rates in buried PVC pipelines

For older water pipeline materials such as cast iron and asbestos cement, future pipe failure rates can be extrapolated from large volumes of existing historical failure data held by water utilities. However, for newer pipeline materials such as polyvinyl chloride (PVC), only limited failure data exists and confident forecasts of future pipe failures cannot be made from historical data alone. To solve this problem, this paper presents a physical probabilistic model, which has been developed to estimate failure rates in buried PVC pipelines as they age. The model assumes that under in-service operating conditions, crack initiation can occur from inherent defects located in the pipe wall. Linear elastic fracture mechanics theory is used to predict the time to brittle fracture for pipes with internal defects subjected to combined internal pressure and soil deflection loading together with through-wall residual stress. To include uncertainty in the failure process, inherent defect size is treated as a stochastic variable, and modelled with an appropriate probability distribution. Microscopic examination of fracture surfaces from field failures in Australian PVC pipes suggests that the 2-parameter Weibull distribution can be applied. Monte Carlo simulation is then used to estimate lifetime probability distributions for pipes with internal defects, subjected to typical operating conditions. As with inherent defect size, the 2-parameter Weibull distribution is shown to be appropriate to model uncertainty in predicted pipe lifetime. The Weibull hazard function for pipe lifetime is then used to estimate the expected failure rate (per pipe length/per year) as a function of pipe age. To validate the model, predicted failure rates are compared to aggregated failure data from 17 UK water utilities obtained from the United Kingdom Water Industry Research (UKWIR) National Mains Failure Database. In the absence of actual operating pressure data in the UKWIR database, typical values from Australian water utilities were assumed to apply. While the physical probabilistic failure model shows good agreement with data recorded by UK water utilities, actual operating pressures from the UK is required to complete the model validation.

A distribution-free lower bound for availability of quantile-based inspection schemes

This paper develops a lower bound for the availability of quantile-based inspection schemes when device lifetimes are assumed to have an increasing failure rate. The lower bound is useful in designing inspection schemes when lifetime distributions are not explicitly known, but must be estimated from limited failure data.

Estimating Weibull Parameters for a General Class of Devices from Limited Failure Data

Relatively simple approaches to estimating Weibull parameters for a general class of devices are developed through regression models. It is assumed that data are collected on a number of device types belonging to a general class. For each device type, the only information available is the number of devices being observed, the total time observed and the total number of failures. By assuming a constant shape parameter and a scale parameter that may vary with the characteristics of the device-type, the least squares method is used to provide estimates of the parameters of a two-parameter Weibull distribution for both replacement and nonreplacement data. An approach is also suggested for dealing with troublesome cases of zero failure occurrences. A numerical example is provided to illustrate the approach.