Abstract
Degradation tests are especially difficult to conduct for items with high reliability. Test costs, caused mainly by prolonged item duration and item destruction costs, establish the necessity of sequential degradation test designs. We propose a methodology that sequentially selects the optimal observation times to measure the degradation, using a convenient rule that maximizes the inference precision and minimizes test costs. In particular our objective is to estimate a quantile of the time to failure distribution, where the degradation process is modelled as a linear model using Bayesian inference. The proposed sequential analysis is based on an index that measures the expected discrepancy between the estimated quantile and its corresponding prediction, using Monte Carlo methods. The procedure was successfully implemented for simulated and real data.
Similar content being viewed by others
References
Broemeling LD (1985) Bayesian analysis of linear models. Dekker, New York
Christen JA, Fox C (2010) A general purpose sampling algorithm for continuous distributions (the twalk). Bayesian Anal 5:263–282
Congdon P (2006) Bayesian statistical modelling, 2nd edn. Wiley, Chichester
Chakraborty S, Gebraeel N, Lawley M, Wan H (2009) Residual-life estimation for components with non-symmetric priors. IIE Trans 41:372–387
Freitas MA, Dos Santos TR, Pires MC, Colosimo EA (2010) A closer look at degradation models: classical and Bayesian approaches. Adv Degrad Mod 3:157–180
Gebraeel N, Lawley MA, Li R, Ryan JK (2005) Residual-life distributions from component degradation signals: a Bayesian approach. IIE Trans 37:543–557
Gelman A, Meng XL, Stern H (1996) Posterior predictive assessment of model fitness via realized discrepancies (with discussion). Stat Sin 6:733–807
Hamada MS, Wilson AG, Reese CS, Martz HE (2008) Bayesian reliability. Springer, New York
Leamer EE (1973) Multicollinearity: a Bayesian interpretation. Rev Econ Stat 55:371–380
Lu CJ, Meeker WQ (1993) Using degradation measures to estimate a time-to-failure distribution. Technometrics 34:161–174
Meeker WQ, Escobar LA (1998) Statistical methods for reliability data. Wiley, New York
Nelson W (1990) Accelerated testing. John Wiley & Sons, Inc., Hoboken, NJ
Peña de la Mora E (2003) Desarrollo de una laca para recubrir acero inoxidable. Master thesis, Centro de Investigació n en Matemáticas-MABE
Rodríguez-Narciso S (2013) Análsis secuencial bayesiano para pruebas de degradación. Ph.D. thesis, Centro de Investigaci ón en Matemáticas
Rodríguez-Narciso S, Villa ER (2011) Stopping times determination for linear degradation process. Estadística 63:77–94
Tsai CC, Tseng ST, Balakrishnan N (2012) Optimal design for degradation tests based on gamma processes with random effects. IEEE Trans Reliab 61:604–613
Tseng ST, Yu HF (1997) A termination rule for degradation experiments. IEEE Trans Reliab 46:130–133
Yu HF, Chiao C (2002) An optimal designed degradation experiment for reliability improvement. IEEE Trans Reliab 51:427–433
Yu HF, Tseng ST (1998) On-line procedure for terminating an accelerated degradation test. Stat Sin 8:207–220
Yu HF, Tseng ST (1999) Designing a degradation experiment. Nav Res Log 46:689–706
Acknowledgments
Part of this research was founded by Ciencia Básica CONACYT research Grant 128477-F. The authors thank the anonymous referees for their corrections and very positive suggestions leading to the improvement of the original manuscript.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: Sensitivity analysis
Here we perform a sensitivity analysis for our sequential method in two perspectives, (a) by modifying the hyperparameters of the prior distribution and (b) by increasing the maximum experiment duration time \(t_c\). In both cases we used synthetic data as in Sect. 4.1, always obtaining reasonable robust results, both in the resulting sequential observation times and the final estimates.
Beyond the examples outlined here, we also carried out several other similar tests, both modifying the priors and the maximum observation time \(t_c\), and with other data sets, always obtaining reasonable robust outputs (results not shown).
1.1 Modifying the hyperparameters
To perform our sensitivity analysis with respect to the prior distribution chosen for the parameters in our model, our sequential analysis was conducted using the same true parameter values and other settings as in Sect. 4.1 (synthetic data), but considering altered hyperparameters values. We present here the same hyperparameter values as used in Sect. 4.1 and an altered set of hyperparameters; see rows 1 and 2 of Table 5, respectively. These altered hyperparameter values follow the guidelines and restrictions established in Sect. 2.1 . We increased \(b_{0}\) and \(b_{1}\) 10 % from row 1 and increased considerably the variances for parameters \(\beta _{0}\), \(\beta _{1},\) \( \theta _{0}^{j},\) and \(\theta _{1}^{j},\) (\(v_0\) increased 32 % and \(v_1\) increased 133 %). Moreover, note that now there is no correlation between \( \beta _{0}\) and \(\beta _{1}\) nor among \(\theta _{0}^{j}\) and \(\theta _{1}^{j}\) since \(\nu _{01}\) and \(u_{01}\) are zero.
The results obtained are summarized in Table 6. The optimum new observation time values are close to the former values. In fact, the third observation time remains exactly the same while the fourth and fifth observation times increased only slightly. We see that these results suggest robustness of our model to prior selection.
1.2 Modifying \(t_{c}\)
In this case we also took the same setting as in Sect. 4.1, with the only modification that \(v_{01}=u_{01}=0\), and changing \(t_{c}=4.5\) and \(t_{c}=5.5\), the maximum observation time. Results are shown in Table 7 and vary slightly within such table and also with respect to Table 6. This suggests that our method also shows robustness to the selection of the maximum experimental time \(t_c\). Note, however, that \(t_c\) cannot be arbitrarily large beyond unit failure unreasonably increasing experimental costs.
1.3 A simulation study for the precision
To address the issue of the precision of the results obtained by our optimal designs we proceed as follows. We simulated two sets of 50 degradation paths under the following conditions
and \(\varepsilon _{ij}\sim \mathrm {i.i.d.N}\left( 0,0.04\right) .\)
Using these data and under the same hyperparameter settings, we conducted the optimal sequential designs. The results are presented in Table 8 and Fig. 7. The resulting designs are basically the same and the sequential analysis is very similar apart from negligible variations. We performed other repetitions, even down to 20 degradation paths, always obtaining very consistent designs (results not shown). Therefore, precise results should be expected by our sequential analysis.
Appendix 2: Goodness of fit
There are some Bayesian goodness of fit tests developed recently in the literature. We choose to use the one proposed by Gelman et al. (1996). All Bayesian goodness of fit approaches are based on the idea that predicted statistics should be coherent with their observations. We use the 0.2 and 0.8 quantiles denoted by \(t_{0.2}\left( \mathbf {y}\right) \) and \(t_{0.8}\left( \mathbf {y}\right) \), respectively, as the test statistics. We generated a large number of J data replicates \(\mathbf {y}^{rep}\) using the posterior predictive distribution obtaining the quantiles \(t_{0.2}\left( \mathbf {y}^{rep}\right) \) and \(t_{0.8}\left( \mathbf {y}^{rep}\right) \) in each replicated sample. We then compare with the observed quantiles to obtain the posterior predicted p value \(p_{q}=\frac{\#\left( t_{q}\left( \mathbf {y}^{rep}\right) >t_{q}\left( \mathbf {y}\right) \right) }{J}\) (\( q=0.2,0.8\)), see Gelman et al. (1996). As a rule of thumb a p value of 0.5 is ideal since this indicates that the observed statistic is in the middle of the posterior predicted distribution.
We evaluated the goodness of fit of the linear model obtained for the lacquer data shown in Fig. 6. We simulated \(J=15,000\) data replicates obtaining the distribution for the posterior predicted quantiles \( t_{0.2}\left( \mathbf {y}^{rep}\right) \) and \(t_{0.8}\left( \mathbf {y} ^{rep}\right) \) shown in Fig. 8. The observed quantiles are 186.6376 and 243.234 respectively. These are shown in Fig. 8 as black vertical lines both nicely placed near the middle of the posterior predicted distributions. The obtained p values are \(p_{0.2}=0.375\) and \( p_{0.8}=0.45\) which are not extreme values showing evidence of an adequate goodness of fit.
Rights and permissions
About this article
Cite this article
Rodríguez-Narciso, S., Christen, J.A. Optimal sequential Bayesian analysis for degradation tests. Lifetime Data Anal 22, 405–428 (2016). https://doi.org/10.1007/s10985-015-9339-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10985-015-9339-7