決定加速預燒試驗的最佳時間和溫度條件

Determine the Optimal Time and Temperature Level of Accelerated Burn-in Test

林揚承(Yang-Chen Lin) ； 李佩熹(Pei-His Lee) ； 童超塵(Chau-Chen Torng)

《品質學報》 16卷4期 (2009/08) Pp. 281-290

摘要

預燒試驗已經廣泛的運用在剔除初期失效率高的電子產品，以避免不良品落入顧客手中。提高溫度能縮短預燒的時間，通常此方法稱之爲加速預燒試驗。當預燒的應力水準不同時，預燒的時間也會隨之改變，Arrhenius-weibull分配能用來表示在加速水準之下具有weibull特徵的電子產品壽命，本研究以Arrhenius-weibull分配來建構成本模式和平均殘餘壽命，並運用到加速預燒的決策模式中，再以基因演算法求解最佳加速預燒時間和溫度水準。最後選用一個TFT-液晶顯示器模組作爲求解的案例說明，敏感度分析說明本研究的模式參數設定對於求解最佳的預燒條件的影響。

關鍵字

加速預燒測試 ； Arrhenius-weibull分配 ； 平均殘餘壽命 ； 基因演算法 ； TFT-液晶顯示器模組

並列摘要

Burn-in has been widely used as an effective procedure for screening out failures in the early operating period of the electronic product, before it is shipped to the customers. Increasing the environmental stress such as temperature can effectively shorten the burn-in time, and this method is usually called accelerated burn-in test. When different stress levels are chosen in the burn-in operation, the burn-in times have to be redetermined. Arrhenius-weibull distribution can describe the Weibull lifetime of electronic products under different temperature levels. In this paper, Arrhenius-weibull distribution and its mean residual life function are applied into the accelerated burn-in cost model, and the genetic algorithms are used to solve the optimal accelerated burn-in time and temperature level. We then choose a real case of TFT-LCD module as an example to illustrate the determination of its optimal accelerated burn-in time and temperature level. A sensitivity analysis of TFT-LCD module case shows the effect of model parameters on optimal burn-in time and temperature level.

並列關鍵字

accelerated burn-in test ； Arrhenius-weibull distribution ； genetic algorithms ； TFT-LCD module

參考文獻

Beneke, M.,Leemis, L. M.(1990).Burn in models and methods: a review.IIE Transactions.22(2),172-180.

Chang, D. S.(2000).Optimal burn-in decision for products with an unimodal failure rate function.European Journal of Operational Research.126(3),534-540.

Chou, K.,Tang, K.(1992).Burn-in time and estimation of change-point with weibull-exponential mixture distribution.Decision Sciences.23(4),973-990.

Jensen, F.,Peterson, N. E.(1982).Burn-In.New York:John Wiley & Sons.

Kim, K. O.,Kuo, W.,Luo, W.(2004).A relation model of gate oxide yield and reliability.Microelectronics Reliability.44(3),425-434.