在現今，產品大多具有高可靠度性質，生產者為了獲得產品的可靠度訊息，如產品壽命的第 ? 百分位數，需要對產品進行實驗進而推估產品的可靠度資訊。加速衰變試驗 (accelerated degradation test, ADT) 則是常被採用的分析工具。為了規劃高效率的加速衰變試驗，尤其是在給定總測試樣本下，如何配置樣本比例至不同加速變數 (如壓力、溫度) 水準上，是本研究所要探討之議題。本文將在加速 Gamma 衰變模型下，分別考量固定效應衰變模型與隨機效應衰變模型之最適實驗設計議題。文獻上，針對兩應力水準的加速衰變試驗樣本數配置問題，多半於高、低兩應力之量測間隔時間與量測次數相同下進行探討。然而實際上大多數加速衰變資料，其高、低應力量測間隔時間往往不相同，因此本研究容許高、低應力量測間隔時間為不相同情境下，探討最適樣本數配置問題，並在不同最適配置準則下，獲得其理論解或數值解結果，也進一步討論其最佳解之漸進性質。結果顯示衰變試驗的量測間隔時間大小確實會對最適樣本數配置造成影響，且不同的準則對於高、低應力最適樣本數的配置策略也有所差異。最後，我們也利用一組實際資料在不同衰變模型或最適配置準則下，分別求得其最適樣本數配置，並進行比較分析，最後透過模擬研究也驗證本文所提之樣本數與應力水準的設置為最佳設計。

The accelerated degradation test (ADT) is widely used to assess the lifetime information of highly reliable products with quality characteristics. To design an efficient ADT plan, the problem of optimizing ADT allocation (including the determinations of stress levels, total testing times, and sample size allocations) is worthy to be discussed. This study assumes that the degradation path follows an accelerated gamma degradation model. Both fixed effect degradation models and random effect degradation models are considered. We allow for different measurement frequencies between high and low stress levels and explore the optimal sample allocation problem. Theoretical or numerical solutions are obtained under various optimal design criteria, and the asymptotic properties of the optimal solutions are further discussed.The results demonstrate that the magnitude of measurement frequencies in degradation tests indeed affects the optimal sample allocation, and different criteria lead to different allocation strategies for high and low stress levels. Finally, using the real data, we determine the optimal sample allocation under different degradation models and optimal design criteria. Through simulation studies, we validate that the sample size and stress level settings proposed in this paper represent the optimal design.

[1] Alzer, H. (1997). On some inequalities for the gamma and psi functions, Mathematics of computation 66(217): 373–389.
[2] Duan, F. and Wang, G. (2019). Optimal design for constant-stress accelerated degradation test based on gamma process, Communications in Statistics-Theory and Methods 48(9): 2229–2253.
[3] Hu, L., Li, L. and Hu, Q. (2017). Degradation modeling, analysis, and applications on lifetime prediction, Statistical modeling for degradation data pp. 43–66.
[4] Kiefer, J. (1974). General equivalence theory for optimum designs (approximate theory), The annals of Statistics pp. 849–879.
[5] Lawless, J. and Crowder, M. (2004). Covariates and random effects in a gamma process model with application to degradation and failure, Lifetime data analysis 10: 213–227.
[6] Lee, M.-Y., Hu, C.-H. and Tang, J. (2017). A two-stage latent variable estimation procedure for time-censored accelerated degradation tests, IEEE Transactions on Reliability 66(4): 1266–1279.
[7] Lim, H. (2015). Optimum accelerated degradation tests for the gamma degradation process case under the constraint of total cost, Entropy 17(5): 2556–2572.
[8] Lim, H. and Yum, B.-J. (2011). Optimal design of accelerated degradation tests based on wiener process models, Journal of Applied Statistics 38(2): 309–325.
[9] Meeker, W. Q., Escobar, L. A. and Pascual, F. G. (2022). Statistical methods for reliability data, John Wiley & Sons.
[10] Pan, Z. and Sun, Q. (2014). Optimal design for step-stress accelerated degradation test with multiple performance characteristics based on gamma processes, Communications in Statistics-Simulation and Computation 43(2): 298–314.
[11] Peng, C.-Y. (2015). Inverse gaussian processes with random effects and explanatory variables for degradation data, Technometrics 57(1): 100–111.
[12] Qi, F. (2020). Some properties of several functions involving polygamma functions and originating from the sectional curvature of the beta manifold, Sao Paulo Journal of Mathematical Sciences 14(2): 614–630.
[13] Shat, H. and Schwabe, R. (2022). Optimal stress levels in accelerated degradation testing for various degradation models, Mindful Topics on Risk Analysis and Design of Experiments: Selected contributions from ICRA8, Vienna 2019, Springer, pp. 113–134.
[14] Tsai, C.-C., Tseng, S.-T. and Balakrishnan, N. (2012). Optimal design for degradation tests based on gamma processes with random effects, IEEE Transactions on Reliability 61(2): 604–613.
[15] Tseng, S.-T., Balakrishnan, N. and Tsai, C.-C. (2009). Optimal step-stress accelerated degradation test plan for gamma degradation processes, IEEE Transactions on Reliability 58(4): 611–618.
[16] Tseng, S.-T. and Lee, I.-C. (2016). Optimum allocation rule for accelerated degradation tests with a class of exponential-dispersion degradation models, Technometrics 58(2): 244–254.
[17] Tung, H.-P., Lee, I.-C. and Tseng, S.-T. (2022). Analytical approach for designing accelerated degradation tests under an exponential dispersion model, Journal of Statistical Planning and Inference 218: 73–84.
[18] Wang, X. and Xu, D. (2010). An inverse gaussian process model for degradation data, Technometrics 52(2): 188–197.
[19] Whittle, P. (1973). Some general points in the theory of optimal experimental design, Journal of the Royal Statistical Society: Series B (Methodological) 35(1): 123–130.
[20] Xu, A.-M. and Cen, Z.-D. (2020). Qi’s conjectures on completely monotonic degrees of remainders of asymptotic formulas of di-and trigamma functions, Journal of Inequalities and Applications 2020: 1–10.
[21] Yang, G. (2007). Life cycle reliability engineering, John Wiley & Sons.
[22] Ye, Z.-S., Chen, L.-P., Tang, L. C. and Xie, M. (2014). Accelerated degradation test planning using the inverse gaussian process, IEEE Transactions on Reliability 63(3): 750–763.