預測一個新產品的失效數量在產品開發過程中是非常重要。在電子產品開發過程中，若能事先知道可能發生的失效數量，則有許多優點，如能準確地估計測試成本和安排適當的人力及測試進度，以及改善產品品質。準確的預測，也可有助於減少測試時間，引導產品準時量產上市及達到更高的品質及可靠度。
本文提出了一個統計模型針對新型筆記型電腦於產品於開發階段的測試過程中可能發現的失效數量及失效累積曲線做預測。本文的研究，係蒐集過去已經開發完成的筆記型產品測試數據，利用穩健回歸(Robust Regression)建立一個失效數量預測模型;本研究發現失效成長則以韋伯(Weibull)分佈模型較其他分佈為佳，可用來預估開發階段過程中隨著時間所發現的失效數量累積。
利用過去的歷史數據來建立一個統計模型，做為目前和未來的新產品的失效數量及趨勢做預測。這樣的預測在測試過程中，可以對產品品質及測試人力資源安排提供了非常有價值的訊息。
研究的驗證結果顯示，這個預測模型用於正在開發的新產品驗證上，具有很精準的失效數量預估，且實際的失效成長曲線也與預估的模型具有高度的相關性。

The defect prediction of a new product is an important task during the development phase in the computer industry. This study presents a new method for predicting the number of defects for new products such as laptops during the development phase.
This approach is based on the robust regression and reliability growth models. Here, the robust regression model is used to predicting the number of defects for new products and the reliability growth model is used to tracking the test progress during the development phase. We found that the Weibull growth model is the best fit compared with other models such as Exponential, three-parameter Logistic and Gompertz growth models.
The prediction defects for four illustrative products are very close to the actual defects during the whole development phase. Also, the Weibull growth curve provides a good fit for monitoring the defect trend.

[1] Card, D. N., “Managing software quality with defects” , Proceeding of Computing Software Application Conference, pp. 472–474 (2002).
[2] Causton, D. R., Venus, J. C. The Biometry of plant Growth, Edward Arnold, London, (1981).
[3] Fenton, N., Neil, M., “A critique of software defect prediction models” , IEEE Transactions on Software Engineering, Vol. 25, pp. 675–689 (1999).
[4] Hampel, F. R., “A general definition of qualitative robustness” , The Annals of Mathematical Statistics, Vol. 42, pp. 1887–1896 (1971).
[5] Hampel, F. R., “The influence curve and its role in robust estimation” , The Annals of Statistics, Vol. 69, pp. 383–393 (1974).
[6] Hampel, F. R., Ronchetti, M., Rousseeuw, P. J., Stahel, W. A., Robust Statistics: The Approach Based on Influence Functions, Wiley, NY, (1986).
[7] Holland, “P. W., Welsch, R. E., “Robust regression using iteratively reweighted least-squares” , Communications in Statistics - Theory and Methods, Vol. 6, pp. 813-827 (1977).
[8] Huang, C. Y., Kuo, S. Y., Chen, I. Y., “Analysis of a software reliability growth model with logistic testing-effort function” , Proceeding of 8th Int. Symp. Software Reliability Engineering, pp. 378–388 (1997).
[9] Huber, P. J., “A robust version of the probability ratio test” , The Annals of Mathematical Statistics, Vol. 36, pp 1753-1758 (1965).
[10] Huber, P. J., “The behavior of maximum likelihood estimates under nonstandard conditions” , Proceedings of the Fifth Berkeley Symposium on Mathematics and Statistics Probability, University of California Press 1, pp. 221–233 (1967).
[11] Huber, P. J., Robust Statistics, Wiley, NY, (1981).
[12] Kyriakides, E., Suryanarayanan, S., Heydt, G., “State estimation in power engineering using the Huber robust regression technique” , IEEE Transactions on Power Systems, Vol. 20, pp. 1183-1184 (2005).
[13] McConnell, S., “Software quality at top speed” , Softw. Develop., Vol. 4, pp. 38-42 (1996).
[14] Nash, J. C., Walker-Smith, M., Nonlinear Parameter Estimation: An Integrated System in BASIC. Marcel Dekker, NY, (1987).
[15] Richards, F. J., “The quantitative analysis of plant growth” , In: Plant Physiology, A Treatise, Vol. VA: Analysis of Growth Behavior of Plants and Their Organs. (Ed.): F.C. Steward, pp. 3-76. Academic Pres, NY, (1969).
[16] Rousseeuw, P. J., Leroy, A. M., Robust Regression and Outlier Detection, Wiley, NY, (1987).
[17] Sandland, R. L., “Mathematics and the growth of organisms-some historical
Impressions” , Mathematical Scientist, Vol. 8, pp. 11-30 (1983).
[18] Sandland, R. L., McGilchrist, C. A. “Stochastic growth curve analysis” , Biometrics, Vol. 35, pp. 255-271 (1979).
[19] Seber, G.A.F., Multivariate Observations, John Wiley & Sons, NY, (1984).
[20] Seber, G.A.F., Wild, C.J., Nonlinear Regression, John Wiley & Sons, NY, (1989).
[21] Staron, M., Meding, W., Soderqvist, B., “A method for forecasting defect backlog in large streamline software development products and its industrial evaluation” , Information and Software Technology, Vol. 52, pp. 1069-1079 (2010).
[22] Staudte, R. G., Sheather, S. J., Robust Estimation and Testing, Wiley, NY, (1990).
[23] Stigler, S., Newcomb, S,. Daniell, P., “The history of robust estimation 1885–1920” , Journal of the American Statistics Association, Vol. 68, pp. 872–879 (1973).
[24] Tukey, J. W,. “The future of data analysis” , The Annals of Mathematical Statistics, Vol. 33, pp. 1–67 (1962).
[25] Wright, S., “Book review” , Journal of the American Statistics Association, Vol. 21, pp 493-497 (1926).