Statistical Inferences for Future Outcomes with Applications to Maintenance and Reliability (original) (raw)
Related papers
Series on Quality, Reliability and Engineering Statistics, 2005
Consider the competing risks situation for a component which may be subject to either a failure or a preventive maintenance action, where the latter will prevent the failure. It is then reasonable to expect a dependence between the failure mechanism and the PM regime. The chapter reconsiders the so called repair alert model which is constructed for handling such cases. A main ingredient here is the repair alert function which characterizes the "alertness" of the maintenance crew. The main emphasis of the chapter is on statistical inference for the model, based on possibly right censored data. Both nonparametric and parametric inference is studied. The methods are applied to two different data sets.
PHM Society Asia-Pacific Conference
Regressive Remaining Useful Life Prediction and Survival Analysis are two lines of research with similar goals but different origins; one from engineering and the other from survival study in clinical research. Although the two research paths share a common objective of predicting the time to an event, researchers from each path typically do not compare their methods with methods from the other direction. Given the mentioned gap, we propose a framework to compare methods from the two lines of research using run-to-failure datasets. Then by utilizing the proposed framework, we compare six models incorporating three widely recognized degradation models along with two learning algorithms. The first dataset used in this study is C-MAPSS which includes simulation data from aircraft turbofan engines. The second dataset is real-world data from streamed condition monitoring of turbocharger devices installed on a fleet of Volvo trucks.
2005
Consider the competing risks situation for a component which may be subject to either a failure or a preventive maintenance action, where the latter will prevent the failure. It is then reasonable to expect a dependence between the failure mechanism and the PM regime. The chapter reconsiders the so called repair alert model which is constructed for handling such cases. A main ingredient here is the repair alert function which characterizes the "alertness" of the maintenance crew. The main emphasis of the chapter is on statistical inference for the model, based on possibly right censored data. Both nonparametric and parametric inference is studied. The methods are applied to two different data sets.
Approximate Prediction Interval for Component Life Based on System Failure Data
Calcutta Statistical Association Bulletin, 1997
Here we consider a two-component system where the components are identical but do not work independently within the system. Assuming the joint distribution of component lives to be Bivariate Exponential (BVED), system (series, parallel or stand-by) life distribution comes out to be a proper mixture of x2-distributions. We approximate the system life distribution as a chi-square with a multiplier that depends on system configuration. Prediction intervals for a single future observation (on component life) are constructed on the basis of system life observations. Further, we have taken Gamma and Rectangular priors for the parameters of the BVED and have approximated the unconditional distribution of system life as Beta with suitable parameters. Taking logarithmic transformation, we get Bayesian prediction intervals.
Reliability Engineering & System Safety, 2013
We look at different prognostic approaches and the way of quantifying confidence in equipment Remaining Useful Life (RUL) prediction. More specifically, we consider: 1) a particle filtering scheme, based on a physics-based model of the degradation process; 2) a bootstrapped ensemble of empirical models trained on a set of degradation observations measured on equipments similar to the one of interest; 3) a bootstrapped ensemble of empirical models trained on a sequence of past degradation observations from the equipment of interest only. The ability of these three approaches in providing measures of confidence for the RUL predictions is evaluated in the context of a simulated case study of interest in the nuclear power generation industry and concerning turbine blades affected by developing creeps. The main contribution of the work is the critical investigation of the capabilities of different prognostic approaches to deal with various sources of uncertainty in the RUL prediction.
Industrial lifetime testing is one of the key procedures for industrial engineers to assess the quality of products or materials. Reliability analysis is hampered by data unavailabilities resulting from multiple failure types, with only the first occurring failure being observable. This leads to major uncertainties about the fitted failure probabilities unless the model satisfies some often hardly verifiable parametric restrictions. This paper contributes to the reliability literature by showing that state-of-the-art statistical models under weak parametric assumptions give informative estimates of failure probabilities. We introduce a new semiparametric bootstrap-based model selection test that allows for testing the validity of these restrictions. Our approach supports the engineer in crafting a parametric model based on data that gives informative results. An empirical analysis of aircraft radio lifetimes demonstrates the estimation of critical model components under various model specifications. The model selection test guides the engineer to select the model with the best fit. We illustrate the practical relevance of data-driven bias reduction techniques for models with dependent censoring.
Uncertainty Quantification in Remaining Useful Life Prediction Using First-Order Reliability Methods
IEEE Transactions on Reliability, 2014
This paper investigates the use of the inverse first-order reliability method (inverse-FORM) to quantify the uncertainty in the remaining useful life (RUL) of aerospace components. The prediction of remaining useful life is an integral part of system health prognosis, and directly helps in online health monitoring and decision-making. However, the prediction of remaining useful life is affected by several sources of uncertainty, and therefore it is necessary to quantify the uncertainty in the remaining useful life prediction. While system parameter uncertainty and physical variability can be easily included in inverse-FORM, this paper extends the methodology to include: (1) future loading uncertainty, (2) process noise; and (3) uncertainty in the state estimate. The inverse-FORM method has been used in this paper to (1) quickly obtain probability bounds on the remaining useful life prediction; and (2) calculate the entire probability distribution of remaining useful life prediction, and the results are verified against Monte Carlo sampling. The proposed methodology is illustrated using a numerical example.
Remaining useful life estimation of critical components based on Bayesian Approaches
2014
Constructing prognostics models rely upon understanding the degradation process of the monitoredcritical components to correctly estimate the remaining useful life (RUL). Traditionally, a degradationprocess is represented in the form of physical or experts models. Such models require extensiveexperimentation and verification that are not always feasible in practice. Another approach that buildsup knowledge about the system degradation over time from component sensor data is known as datadriven. Data driven models require that sufficient historical data have been collected.In this work, a two phases data driven method for RUL prediction is presented. In the offline phase, theproposed method builds on finding variables that contain information about the degradation behaviorusing unsupervised variable selection method. Different health indicators (HI) are constructed fromthe selected variables, which represent the degradation as a function of time, and saved in the offlinedatabase as r...
Annual Conference of the PHM Society, 2019
Estimating accurate Time-of-Failure (ToF) of a system is key in making the decisions that impact operational safety and optimize cost. In this context, it is interesting to note that different approaches have been explored to tackle the problem of estimating ToF. The difference is in part characterized by different definitions of the hazard zones. The conventional definition for the cumulative distribution function (CDF) calculation is assumed to have well-defined hazard zones, that is, hazard zones defined as a function of the system state trajectory. An alternate method suggests the use of hazard zones defined as a function of the system state at time , instead of hazard zones defined as a function of system state up to and including time k (Acuña and Orchard 2018, 2017). This paper explores these differences and their impact on ToF estimation. Results for the conventional CDF definition indicated that, (i) the cumulative distribution function is always an increasing function of t...
Reliability Engineering & System Safety, 2009
Many times, reliability studies rely on false premises such as independent and identically distributed time between failures assumption (renewal process). This can lead to erroneous model selection for the time to failure of a particular component or system, which can in turn lead to wrong conclusions and decisions. A strong statistical focus, a lack of a systematic approach and sometimes inadequate theoretical background seem to have made it difficult for maintenance analysts to adopt the necessary stage of data testing before the selection of a suitable model. In this paper, a framework for model selection to represent the failure process for a component or system is presented, based on a review of available trend tests. The paper focuses only on single-time-variable models and is primarily directed to analysts responsible for reliability analyses in an industrial maintenance environment. The model selection framework is directed towards the discrimination between the use of statistical distributions to represent the time to failure (''renewal approach''); and the use of stochastic point processes (''repairable systems approach''), when there may be the presence of system ageing or reliability growth. An illustrative example based on failure data from a fleet of backhoes is included.