在許多可靠度分析及壽命研究中常會因為時間、人力和金錢的限制或成本的考量而無法取得完整的樣本資料,而在壽命檢測的實驗中,各種設限的方法都是為了節省時間與成本,一般化型二逐步設限就是其中的一種。 Gompertz分配是由Gompertz在1825年所提出的,此模型是個非常重要的生長模型,也運用在生物、醫學、精算等研究上。 極值分配的應用廣泛,運用於自然現象的降雨量及土壤,也運用於壽命模型。 在本文中我們所要探討的主題就是在隨機移除之一般化型二逐步設限的方法下所獲得之有序樣本經過適當的轉換,可得到一組來自標準指數分配且互相獨立的樣本,接著利用此組樣本形成不同樞紐量,和我們所提出的一組新的樞紐量,對來自於雙參數Gompertz分配和雙參數極值分配的形狀參數做假設檢定和信賴區間,以及建立雙參數之聯合信賴區域。最後利用電腦模擬的方法,在不同的(n,m,p,r)下找出檢定力最好以及信賴區間長度最短和聯合信賴區域面積最小的樞紐量。 In many lifetimes analysis, we can’t collect the whole data due to the restriction of time, cost and material, censoring arises. There are several types of censoring schemes and the General Type II Progressive censoring scheme is one of those. The Gompertz distribution is one of the most important growth models. The Gompertz mortality function was established by Gompertz (1825) and it has many applications in, biological, medical, and actuarial studies.The Extreme-value distribution has been extensively used to model natural phenomena such as rainfall and floods, and also in modeling lifetimes. Therefore, the interval estimations of two parameters and the hypothesis testing under General Type II progressive censoring with random removals are proposed in this research. We proposed a set of new pivotal quantities with new distribution to be compared with some other pivotal quantities with F distribution in this paper.