An Evaluation Method of Accelerated Storage Life Test for Typical Mechanical and Electronic Products

Reliability of long-term storage products is related to
the availability of the whole system, and the evaluation of storage life
is of great necessity. These products are usually highly reliable and
little failure information can be collected. In this paper, an analytical
method based on data from accelerated storage life test is proposed to
evaluate the reliability index of the long-term storage products. Firstly,
singularities are eliminated by data normalization and residual
analysis. Secondly, with the preprocessed data, the degradation path
model is built to obtain the pseudo life values. Then by life distribution
hypothesis, we can get the estimator of parameters in high stress levels
and verify failure mechanism consistency. Finally, the life distribution
under the normal stress level is extrapolated via the acceleration model
and evaluation of the actual average life is available. An application
example with the camera stabilization device is provided to illustrate
the methodology we proposed.

• Jinyong Yao
• Hongzhi Li
• Chao Du
• Jiao Li

• Accelerated storage life test
• failure mechanism consistency
• life distribution
• reliability.

[1] Wang Z B, Ren W B, Li G F. Review of Accelerated Degradation
Testing and Accelerated Life Testing. Low Voltage Apparatus, 2010,
(1):1-6.
[2] Gao C, Jiang M. Degradation Path Modeling Method Based on Time
Series Analysis. Applied Mechanics and Material, 2012, 3(6):121-126.
[3] Zhong Q H, Zhang Z H, Wang L. About model choice of degradation
data analysis method. Systems engineering, 2009, 27(11): 111-114.
[4] Gopikrishnan A. Reliability inference based on degradation and time to
failure data: some models, methods and efficiency comparisons. The
University of Michigan. 2004.
[5] Meeker W Q, Escobar L A, Lu J C. Accelerated degradation tests:
modeling and analysis (J). Technometrics. 1998, 40(2):89-99.
[6] Ma X B, Wang J Z, Zhao Y. Based on the degradation data for life
distribution reliability assessment method. Systems engineering and
electronics. 2010, 33(2): 261-264 (in Chinese).
[7] Zhao Y T. Electromagnetic valve storage life assessment (J) quality and
reliability, 2010(5): 25-28.
[8] Filzmoser, P.Walczak, B. What can go wrong at the data normalization
step for identification of biomarkers? Journal of Chromatography A. Oct
3, 2014, Vol. 1362, p194, 12 p.
[9] Carrasco, Jalmar MF, Ortega, Edwin MM,Paula, Gilberto A.
Log-modified Weibull regression models with censored data: Sensitivity
and residual analysis. Computational Statistics and Data Analysis, 2008
52(8): 4021-4039.
[10] Kim Seong-Joon, Bae Suk Joo. Cost-effective degradation test plan for a
nonlinear random-coefficients model. In Reliability Engineering and
System Safety February 2013 110: 68-79.
[11] Chaos M T. Degradation analysis and related topics: some thoughts and
a review. ProcNatlSciCounc. ROC(A), 1999, 23(5): 555-566．
[12] Wang Y H, Li X G. Study on Accelerated Storage Life Test of Initiating
Explosive Device Based on Step Stress Method. Equipment
Environmental Engineering. 2013, 10(1):38-40.
[13] Li X J. Study on the Confidence Limits for the Mean of Weibull
Distributions. Chinese Journal of Applied Probability and Statistics.