Weibull Distribution Part-1

Dear viewers, we are happy to release this 25th video from Institute of Quality and Reliability! This is the first of our two videos on Weibull Distribution. In this video, Hemant Urdhwareshe explains the basic concepts of Weibull Distribution, its cumulative and probability functions, hazard function and effect of shape parameter on these functions. Hemant also explains the related mathematical treatment briefly. The concepts are explained with an application case study of tires to calculate reliability within warranty period. The video additionally explains calculating probability of failure and reliability using Microsoft Excel functions. The videos would be useful to those who want to learn applicability of Weibull Distribution to estimate reliability and to those who wish to take ASQ CRE, CQE and CMBB certification exams.

[@cincyaadi]

This is the best thing that I came across while studying Weibull thanks. Visualization and Derivation helped a lot.

[@instituteofqualityandrelia7902]

Thank you for your feedback! Appreciate!

[@fernandomatosinhos8583]

Fantastic class!! I am watching from Brazil...Your explanation calmly developing practical exercises it is a fantastic way to clarify to viewers how to use the formulations. Congratulations, dear professor.

[@instituteofqualityandrelia7902]

Hello Fernando! Thnak you and I am glad it was helpful!

[@youngmlee3001]

Excellent explanation of Weibull distribution and reliability analysis. Thanks!!!

[@instituteofqualityandrelia7902]

Glad it was helpful!

[@ashishshrivastava8864]

Thank you for your videos. It's very helpful for many of us.

[@instituteofqualityandrelia7902]

Welcome. My pleasure!

[@TheSanthoshteja]

Thanks for considering and making video on Weibull sir.

[@uhemant1]

Welcome! Your feedback is welcome!

[@eamona708]

Intro was straight fire!!

[@instituteofqualityandrelia7902]

Interesting! Did you find it useful?

[@alberttang8388]

Excellent!Very detailed!

[@instituteofqualityandrelia7902]

Thank you!

[@justanotherchannel4271]

Watching from Ireland - very helpful demonstration!

[@uhemant1]

My pleasure! Thank you!

[@chikaijoma9134]

Weibull made easy. Great video.

[@instituteofqualityandrelia7902]

Glad it helped!

[@AC-ll5ky]

Very good presentation! Thx you Sir

[@instituteofqualityandrelia7902]

Most welcome!

[@billyjames6407]

Great video, thank you.

[@instituteofqualityandrelia7902]

Glad you liked it!

[@DoWeKnow_World]

Superb explanation with illustration..

[@instituteofqualityandrelia7902]

Thanks a lot 😊

[@sunnyj1967]

Awewsome presenatation.

[@uhemant1]

Thank you!

[@mahendrakanchan4715]

Good technical statistics...

[@instituteofqualityandrelia7902]

Thanks a lot for appreciation!

[@kerguule]

Hello! If I know the weibull parameters, I can calculate the MTBF using the gamma function and thus the failure rate. How can I calculate MTBF and FR so that the corrective maintenance (e.g. with 100h intervals) is taken into account and thus the FR is lower? Which function takes these into account? Excellent videos, keep up the work!

[@instituteofqualityandrelia7902]

My apologies for late response! The Weibull Distribution is appropriate for a single failure mode in non repairable systems. So why would you need MTBF in such case?

[@kerguule]

@@instituteofqualityandrelia7902 Thank you for your response! Now after while I found an equation where one can calculate average failure rate (note the word average) over time: FRavg = ( 1 - R(T) ) / ( ∫ from 0 to T R(t) dt ) where T is the maintenance interval. There is also an approximation for this without the integral which is FRavg = ( -ln R(T) ) / T. I found these from the book called Reliability Theory and Practice by Igor Bazovsky. Do you think these are good enough for rough FR estimations or should one use some sort of Monte Carlo simulation with the power law models etc?

[@kerguule]

@@instituteofqualityandrelia7902 Just one more thing. Another form for the rough approximation equation is with Weibull parameters: FRavg = ( t^(β-1 ) ) / ( η^β ) where β is the shape and η is the scale parameter. This should give exact the same results as with using the reliability R(T) in the equation.

[@MohammadBintangRizqiPratama]

Thanks for the explanation, but still i have question. In a real case e.g determining the components reliability of a machine, what is the ß value? Because sometimes, when it broke, u just need to repair the components and it will run again (basically we use MTBF on this situation), and sometimes when it broke you need to change the components so it would start again (non repairable units, and this we use MTTF). What is the ß value then?

[@instituteofqualityandrelia7902]

Beta value is estimated from the data of past failures.

[@solomontekeste5551]

thanks dear...sorry, I need to study the mean and median of survival time from weibull distribution? thanks.

[@instituteofqualityandrelia7902]

Median is just 50th percentile. So its easy. Mean requires use of formula.

[@dinesh-uw2yi]

Sir excellent explination can you tell me how to draw that graphs.

[@instituteofqualityandrelia7902]

Thanks. The graphs can be easily plotted on XL using Weibull functions and formulae.

[@luigi1606]

Ok so basically β is kind of like an inverse standard deviation for failure over time. The higher it is, the less likely is it to have failures far from the assigned t.

[@instituteofqualityandrelia7902]

Interesting interpretation!

[@alaaisraa9459]

Thank you for this video, please I have a question, how do we estimate the failure function, if I have the number of data exactly 20

[@instituteofqualityandrelia7902]

I am not sure whether I understand your question. But you need to calculate median rank using the formula shown (j-0.3)/(n+0.4). For 20 data points, n=20. Alternatively, you can use median rank table of n=20.

[@alaaisraa9459]

@@instituteofqualityandrelia7902 merci monsieur pour ma question comment on va calculer ou bien comment on va estimer la fonction de défaillance ,si notre donnés égale à 20 J'attend votre favorable réponse est on utilise la méthode des rangs médium pour n inférieur à 20 ou on utilise la méthode des rangs moyen quand n est entre 20et 5O ni/n+1 j'attend votre favorable réponse, SVP

[@alaaisraa9459]

@@instituteofqualityandrelia7902 Merci pour ma question comment on va calculer ou bien comment on va estimer la fonction de défaillance ,si notre donnés égale à 20 est on utilise la méthode des rangs médium pour n inférieur à 20 ou on utilise la méthode des rangs moyen quand n est entre 20et 5O ni/n+1 j'attend votre favorable réponse, SVP

[@instituteofqualityandrelia7902]

Unfortunately I am unable to understand your language. Please write in English.

[@judithkibirango2354]

@@instituteofqualityandrelia7902 Hello can you help derive the formula for survival and hazard function for a 2 parameter weibull distribution and also derive the formula for the survival and hazard function for the log logistic distribution

[@shanmugavelayutham1394]

Sir, I have 1 doubt, at 5.45 you said for alpha you mentioned shape parameter. but actually Beta is called as share parameter is it? but in excel we have to consider alpha as a shape parameter. is right?

[@instituteofqualityandrelia7902]

This is only because in XL, the notation is different. In most literature, the shape parameter is denoted as beta. In XL, the shape parameter is denoted as alpha.

[@sumitgharge1298]

Hello sir can you please tell me how we can find the values of shape parameter and scale parameter??

[@instituteofqualityandrelia7902]

Watch my other video: kzbin.info/www/bejne/mqTYfYmJZ96Earcsi=i7saq3ej5S-wkYKk

[@sumitgharge1298]

@@instituteofqualityandrelia7902 Sir can you tell me how we can decide which plots are good and which one is to choose or very helpful for further analysis after conducting the ALT How to decide plots according to what??

[@brahimboutoutou4351]

Why t=5000, in the example 5000 km

[@instituteofqualityandrelia7902]

Kilometer is a unit of time for reliability calculation. Apologise for the late reply. Good luck!

[@mariyamsiddiqui2321]

Can you please tell me the book name based on operation research and reliability.

[@instituteofqualityandrelia7902]

There may not be one single book for these as these are different subjects.

[@rahulhinge4583]

good wishes sir....I am in USA.....wanting to be a reliability engg.....can I get online certificate of your institute....online classes??????....this video is very amazing and simple....

[@instituteofqualityandrelia7902]

Yes you can. You need to take our exam. You need to keep your camera on. with Google Hangout. Please send separate mail to me on [e-mail].

[@rahulhinge4583]

@@instituteofqualityandrelia7902 ok...can I do this after 12 May....that day I will done with my college stuff.....may I know the fees.....?

[@instituteofqualityandrelia7902]

Rahul, we plan to launch online CRE workshop soon!

[@luisfernandomacedo1451]

@@instituteofqualityandrelia7902 Hi, I'm Luis from Brazil. Nice explanation! Will the workshop be presented on You Tube?

[@instituteofqualityandrelia7902]

@@luisfernandomacedo1451 Hello Luis Fernando Macedo, Greetings! The workshop is scheduled from 3-April to 26-April. It will be through Zoom link and not on KZbin. For more information and details, visit this link: www.world-class-quality.com/training-calendar. With best wishes, Hemant

[@arifurmollah4386]

I have a question for you, in colleges and universities, which courses talk about Weibull distribution? Which courses talk about that? Which courses? Another question is which colleges and universities talk about Weibull distribution? Which colleges and universities have that? Which colleges and universities?

[@instituteofqualityandrelia7902]

Thanks. But I have not understood your question.

[@arifurmollah4386]

Which question did you not understood? Which question? Which one?

[@crazykoala2289]

Im a mechanical eng. student and we learn Weibull as part of "Reliability of Systems" course at 4th year bro.

[@arifurmollah4386]

@@crazykoala2289 Thank you for telling me this. Thank you so much for this. 🙂

[@crazykoala2289]

You're welcome bro :) @@arifurmollah4386

[@sukhjeetkaur5156]

not interactive, scale parameter not explained

[@instituteofqualityandrelia7902]

Unfortunately KZbin videos are not interactive, but we do it in our training sessions. I suggest you watch complete series of videos on Weibull distribution and hazard rate.