Degradation Test Plan for a Nonlinear Random-Coefficients Model

Sample size and inspection schedule are essential components in degradation test plan. In practice, an experimenter is required to determine a certain level of trade-off between total resources and precision of the degradation test. This paper develops a design of cost-efficient degradation test plan in the context of a nonlinear random-coefficients model, while satisfying precision constraints for the failure-time distribution derived from the degradation testing data. The test plan introduces a precision metric to identify the information losses due to reduction of test resources, based on the cost function to balance the test plan. In order to determine a cost-efficient inspection schedule, a hybrid genetic algorithm is used to solve a cost optimization problem under test precision constraints. The proposed method is applied to degradation data of plasma display panels (PDPs). Finally, sensitivity analysis is provided to show the robustness of the proposed test plan.

[1]  M. Boulanger,et al.  Experimental Design for a Class of Accelerated Degradation Tests , 1994 .

[2]  É. Moulines,et al.  Convergence of a stochastic approximation version of the EM algorithm , 1999 .

[3]  Suk Joo Bae,et al.  Cost-effective degradation test plan for a nonlinear random-coefficients model , 2013, Reliab. Eng. Syst. Saf..

[4]  Hong-Fwu Yu,et al.  Designing a degradation experiment , 1999 .

[5]  Suk Joo Bae,et al.  A Nonlinear Random-Coefficients Model for Degradation Testing , 2004, Technometrics.

[6]  Suk Joo Bae,et al.  Direct Prediction Methods on Lifetime Distribution of Organic Light-Emitting Diodes From Accelerated Degradation Tests , 2010, IEEE Transactions on Reliability.

[7]  V. B. Melas,et al.  Design and Analysis of Simulation Experiments , 1995 .

[8]  L. Sheiner,et al.  Modelling of individual pharmacokinetics for computer-aided drug dosage. , 1972, Computers and biomedical research, an international journal.

[9]  Shuo-Jye Wu,et al.  Optimal design of degradation tests in presence of cost constraint , 2002, Reliab. Eng. Syst. Saf..

[10]  Alain Mallet,et al.  Optimal design in random-effects regression models , 1997 .

[11]  C. Metzler.  Usefulness of the Two-Compartment Open Model in Pharmacokinetics , 1971 .

[12]  James H. Matis,et al.  Stochastic Models of Compartmental Systems , 1979 .

[13]  D. Bates,et al.  Nonlinear mixed effects models for repeated measures data. , 1990, Biometrics.

[14]  Suk Joo Bae,et al.  Dual Features Functional Support Vector Machines for Fault Detection of Rechargeable Batteries , 2009, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[15]  S. Tseng,et al.  Step-Stress Accelerated Degradation Analysis for Highly Reliable Products , 2000 .

[16]  Hong-Fwu Yu,et al.  Designing a Degradation Experiment with a Reciprocal Weibull Degradation Rate , 2004 .

[17]  William Q. Meeker,et al.  A COMPARISON OF DEGRADATION AND FAILURE-TIME ANALYSIS METHODS FOR ESTIMATING A TIME-TO-FAILURE DISTRIBUTION , 1996 .

[18]  Jun Shao,et al.  RELIABILITY ANALYSIS USING THE LEAST SQUARES METHOD IN NONLINEAR MIXED-EFFECT DEGRADATION MODELS , 1999 .

[19]  John Yen,et al.  A hybrid approach to modeling metabolic systems using a genetic algorithm and simplex method , 1998, IEEE Trans. Syst. Man Cybern. Part B.

[20]  Marc Lavielle,et al.  Maximum likelihood estimation in nonlinear mixed effects models , 2005, Comput. Stat. Data Anal..

[21]  Min Xie,et al.  Classifying Weak, and Strong Components Using ROC Analysis With Application to Burn-In , 2007, IEEE Transactions on Reliability.

[22]  Suk Joo Bae,et al.  Degradation Analysis of Nano-Contamination in Plasma Display Panels , 2008, IEEE Transactions on Reliability.

[23]  C. Joseph Lu,et al.  Using Degradation Measures to Estimate a Time-to-Failure Distribution , 1993 .

[24]  Enrico Zio,et al.  Designing optimal degradation tests via multi-objective genetic algorithms , 2003, Reliab. Eng. Syst. Saf..