If defects are detected from line, is it caused by machine itself or by the whole line?


Significance analysis

Significance table

  The production manager often meets the scenaria that series of machines are linked together forming one line to produce. And then we find the spoilages after being incubated. Do the soilages or unsterilities come from the indiviual machines or fillers such as 1, 2, 3, and 4. Or they come from UHT? 

  The significance table can give you the answer.

   we need to take samples from the doubted filler and the reference. For example, we doubt the defects come from machine 3 and the other machines are normal. When taking samples from Filler 3, we also need to take samples from one reference machine such as 1, 2, or 4.

   If the spoilages are found from machine 3 being 9, and 3 defects are found from filler 1, 2 and 4 at the same time. We can have the conclusion that machine 3 has its own problem besides the common problems from all of the mchines at the risk of 20%. If spoilages from filler 3 are 10 or 12, the rish will be 10% or 5% respectively. 

  We can query the counted unsterilities from different machines to compare. And often we use the risk being 10% to judge.