Hello,

I would like to know that if there are no defects in units used for samples, how could we calculate sigma level for that product?

For example

In a process 10000 files completed in 3 months and there are no errors in this duration in any of the file. Now I have to calculate the sigma level of this process for 3 months, what method I should use?

Please help me.

Thank you

Hi Arik,

Sigma level cannot be calculated in this situation. This is the disadvantage of discrete data. Unless you get defects/defectives, you cannot assess the z value.

Best Regards,

VK

I think one may go for Continuous data instead of Discrete data in these cases.

Continuous data can assume a range of numerical responses on a scale and it can be used to measure a wide range of different values which will lead to even the slightest defect being detected and measured.

Can someone please suggest, if this can be the correct approach?

You are absolutely right Inderjeet. If we have an option we mostly prefer continuous data, especially in a critically important process.

As the others already stated, we are unable to calc a sigma level. Nevertheless, you could use the Bernoulli distr. to quantify your quality. 

I suppose you have a well defined quality goal like "first pass yield (FPY) must be at least XYZ%". If this is true, you can setup a hypothesis test and check whether or not or FPY is at least XYZ% -- just assume that your FPY is exactly XYZ% and calculate the probability that you obtain no defect.