本論文主要探討Logistic-based可靠度模型之實用性。透過數據嵌合得知該模型之嵌合結果良好，再以隨機選取部分數據進行嵌合來檢驗Logistic-based可靠度模型的敏感性以及對完整數據可靠度的重現性。從不完整數據的分布得知數據取樣的重要性並可制定規範，由結果可知，數據分布至反曲點附近已可準確重現完整數據之可靠度，因此當壽命試驗通過反曲點後即可中止數據的收集。以Logistic-based可靠度模型為依據討論複雜系統參數調整影響及後續之內文。This present paper mainly discusses practicality of the Logistic-based reliability model. By result of fitting data knows that the model is accurate, and carries on the fitting by the random selection part of data to examine the sensitivities and reappearance of completed data. Distribution of incomplete data could be catches on important to sampling data and formulate the standard. It could display the reliability of completed data precisely as the data pass through the inflection point. When the life test comes to inflection point may consider stopping collect data. Based on Logistic-based model to discuss the complex system by influence of parameter adjustment and following.
