Gumbel Distribution Homework: Proportion of Switches Failing Spec

  • Thread starter mikeliebelt
  • Start date
In summary, the conversation discusses the proportion of switches that would fail to meet specification for their output rise time, given a certain process capability and mean. The first two parts of the question were solved using the normal distribution and the Gumbel distribution was mentioned as a potential solution for the third part. The standard deviation for the Gumbel distribution was given and calculations were done to find the probability of switches failing to meet the given specification.
  • #1
mikeliebelt
1
0

Homework Statement


I am not very familiar with this distribution.It asking me the proportion of switches that fails to meet specification.The questoin is:

The specification for the output rise time of a bipolar Hall switch is between 0.23 and 0.27 microseconds. Fly Electronics can only achieve a process capability ( ) of 0.8.
(i) If Fly electronics achieve a process mean of 0.25, what proportion of switches would fail to meet specification if rise times are normally distributed?
(ii) Suppose Fly Electronics manufactured switches with a mean that is off centre and only achieve a process performance ( ) of 0.7. What proportion of switches would fail to meet specification if rise times are normally distributed?

(iii) If Fly electronics achieve a process mean of 0.25, what proportion of switches would fail to meet specification if rise times have a Gumbel distribution?

I found out that standard deviation is 8.33*10^-3.I did first 2 parts,not sure how to do 3.





Homework Equations



Gumbel distr. has cdf
F(x)=exp(-exp(-(x-e)/d))
mean=e+0.57722*d
sd=1.28255*d

The Attempt at a Solution

 
Physics news on Phys.org
  • #2
I have done (i) and (ii),not sure how to do (iii).For (i) I did z1=(0.23-0.25)/(8.33*10^-3)= -0.24 and z2=(0.27-0.25)/(8.33*10^-3)= 0.24. Then I used the excel NORMSDIST function to get the probability of each z score,which gave me 0.41 and 0.41 respectively.Then I subtracted one from another, which gave me 0.09.For (ii) I did z1=(0.23-0.25)/(12.5*10^-3)= -0.16 and z2=(0.27-0.25)/(12.5*10^-3)= 0.16. Then I used the excel NORMSDIST function to get the probability of each z score,which gave me 0.54 and 0.54 respectively.Then I subtracted one from another, which gave me 0.08.
 

FAQ: Gumbel Distribution Homework: Proportion of Switches Failing Spec

What is the Gumbel distribution?

The Gumbel distribution is a type of probability distribution used in statistics to model extreme values or events that occur rarely, such as natural disasters or stock market crashes. It is a continuous distribution that is often used in reliability and survival analysis.

How is the Gumbel distribution related to switches failing spec?

The Gumbel distribution can be used to model the failure times of switches, which are electronic components used in electrical systems. By analyzing the proportion of switches failing spec, we can gather information about the reliability of these switches and make predictions about future failures.

What is the significance of the proportion of switches failing spec?

The proportion of switches failing spec is an important metric in the field of reliability engineering. It provides insight into the overall performance and reliability of a system, and can help identify potential issues or areas for improvement.

How is the Gumbel distribution used in analyzing the proportion of switches failing spec?

The Gumbel distribution is often used to model the failure times of switches, as it can accurately capture rare and extreme events. By fitting the Gumbel distribution to a set of failure data, we can estimate the proportion of switches failing spec and also make predictions about future failures.

What are some limitations of using the Gumbel distribution for this analysis?

One limitation of using the Gumbel distribution is that it assumes that the failures are independent and identically distributed, which may not always be the case in real-world scenarios. Additionally, the Gumbel distribution may not be the best fit for all types of failure data, so it is important to consider other distributions and compare their performance as well.

Similar threads

Replies
5
Views
3K
Replies
2
Views
5K
Replies
21
Views
3K
Replies
1
Views
1K
Replies
12
Views
3K
Back
Top