What is the relationship between the Crow AMSAA Model and the Weibull distribution? I've read that the Crow-AMSAA distribution is equivalent to the Weibull Intensity Function. You said in this video that it is represented by the Power law distribution.
All growth models are power law and NHPP. Weibull distribution is applicable to a non-repairable item with a particular failure mode. The intensity function looks like Weibull but is not Weibull distribution.
Dear Hemant, congratulations on the excellent video. Could you tell us the name of the book that presents the formulas used to estimate the constants a and b?
It is 'Introduction to Reliability and Maintainability Engineering ' by Charles Ebeling. Here is a link to buy on Amazon. AN INTRODUCTION TO RELIABILITY AND MAINTAINABILITY ENGINEERING amzn.in/d/064jgdjC
During development, it is nearly impossible that there will be no failure. In such a case, the growth model will not apply. Gowth means reliability improvement. If there are no failures, no question of growth.
How can i use this model/calculation for semiconductor product where we have increased num of samples/hardware with time and progressively with different firmwares, reliability improves ?
@@uhemant1thanks. for an example with Firmware1 with 100 samples for 1 day test duration i got 4 failures then i switched to Firmware 2 and added additional 100 samples , With test duration of 2 days i got 3 failure in these 200 samples. so Genset1 will be Firmware1 and Genset2 will be Firmware2. but all 100 samples will show 1 day duration and 200 samples in firmware 2 will show 2 days test duration. It will not like above example with variable test duraiton right ?
Hello Fernando, Thanks for your interest. There are two gensets. So the time is time column shows data when any of the genset fails. Hope this is clear.
Yes I got it. When performing this analysis using Minitab the results are different. Also, how this Eta and Beta relates to shape and scale parameters informed when using Minitab?
Great job. I've subscribed. Informative and quick to the point.
Thank you! My pleasure!🎉
What is the relationship between the Crow AMSAA Model and the Weibull distribution? I've read that the Crow-AMSAA distribution is equivalent to the Weibull Intensity Function. You said in this video that it is represented by the Power law distribution.
All growth models are power law and NHPP. Weibull distribution is applicable to a non-repairable item with a particular failure mode. The intensity function looks like Weibull but is not Weibull distribution.
Dear Hemant, congratulations on the excellent video. Could you tell us the name of the book that presents the formulas used to estimate the constants a and b?
It is 'Introduction to Reliability and Maintainability Engineering ' by Charles Ebeling. Here is a link to buy on Amazon.
AN INTRODUCTION TO RELIABILITY AND MAINTAINABILITY ENGINEERING amzn.in/d/064jgdjC
Good morning, Hermant! Thank you for attention!
Good afternoon sir
Nice lecture.
With Regards
Welcome. Good to know you liked it!
Thank you Sir
Welcome!
How to do calculation when i have 4 system in place and there is no failure reported at all??
During development, it is nearly impossible that there will be no failure. In such a case, the growth model will not apply. Gowth means reliability improvement. If there are no failures, no question of growth.
How can i use this model/calculation for semiconductor product where we have increased num of samples/hardware with time and progressively with different firmwares, reliability improves ?
I believe you can use the same procedure.
@@uhemant1thanks. for an example with Firmware1 with 100 samples for 1 day test duration i got 4 failures then i switched to Firmware 2 and added additional 100 samples , With test duration of 2 days i got 3 failure in these 200 samples. so Genset1 will be Firmware1 and Genset2 will be Firmware2. but all 100 samples will show 1 day duration and 200 samples in firmware 2 will show 2 days test duration. It will not like above example with variable test duraiton right ?
Why in some cases there is a time in the column even when genset did not fail?
Hello Fernando, Thanks for your interest. There are two gensets. So the time is time column shows data when any of the genset fails. Hope this is clear.
Yes I got it. When performing this analysis using Minitab the results are different. Also, how this Eta and Beta relates to shape and scale parameters informed when using Minitab?
Why you sum de hours information of genset 1 and genset 2?
Because calculation of failure intensity requires total hours.