接著本篇論文還會繼續把良率與隨機性的相關數值(NBD及Bscore)做進一步的整合，以座標的方式找出在座標平面上的點後，找到群集中心後，經由各群集間的點到群集中心的距離作為判斷，找出各自的族群，再加以分析判別各族群的晶圓，以達到快速測試檢定的效果。 ;Testing needs to consume a lot of time and human resources. In this period of rapid technological progress, it is necessary to refine the testing and discover errors and constantly appearing results. The purpose of this paper is to produce throughput. And group into groups to find out the performance of yield or randomness, in order to achieve the effect of rapid classification detection.
There have been papers on the uniformity analysis of wafer yield and randomness. This paper will do more careful discrimination and classification. First, the yield uniformity analysis and random uniformity analysis will be discussed separately. Compare with each other, look for similar ones as the same cluster, and then group out the dissimilar ones, until all ethnic groups are wafers of the same group.
Then this paper will continue to further integrate the relevant values of yield and randomness (NBD and Bscore). After finding the point on the coordinate plane by means of coordinates, after finding the cluster center, go through the The distance from the point to the cluster center is used as a judgment to find out the respective ethnic groups, and then analyze the wafers of each ethnic group to achieve the effect of rapid test verification.
