Confidence level calculation

Post all other topics which do not belong to any other category.
Post Reply
User avatar
Hyperbar00
Inserter
Inserter
Posts: 24
Joined: Sun Jan 28, 2018 6:47 pm
Contact:

Confidence level calculation

Post by Hyperbar00 »

Greetings everyone,

I was looking at the Uranium processing page on the game wiki : https://wiki.factorio.com/Uranium_processing especially the confidence level calculation and I found myself interested by the formulas used to obtain these results.
Could anyone provide them and why not the different steps and values please ?

Thank you in advance

User avatar
DaveMcW
Smart Inserter
Smart Inserter
Posts: 3699
Joined: Tue May 13, 2014 11:06 am
Contact:

Re: Confidence level calculation

Post by DaveMcW »

POISSON.DIST() Excel function

failure_chance = POISSON.DIST(required_number_of_successes, success_chance * number_of_events, TRUE)

failure_chance = POISSON.DIST(40, 0.007 * 7000, TRUE)

failure chance for 7000 cycles = 10.98%

User avatar
DaveMcW
Smart Inserter
Smart Inserter
Posts: 3699
Joined: Tue May 13, 2014 11:06 am
Contact:

Re: Confidence level calculation

Post by DaveMcW »

Looking at the wiki history, the table and graph were obtained by using a simple ratio. That is the wrong way to do it.

User avatar
Hyperbar00
Inserter
Inserter
Posts: 24
Joined: Sun Jan 28, 2018 6:47 pm
Contact:

Re: Confidence level calculation

Post by Hyperbar00 »

Thanks A LOT !

I was trying to use Binomial Distribution and quickly noticed that it wasn't the right way because my result ended being undefined upon reaching high values.
So I couldn't figure how to get the job done until you gave me the answer. A huge thank you :D

I was looking for this formula because of the Realistic Fusion mod which gives a 0.02% chance to get 1 Deuterium so I wanted to get some probabilities and stuff in the same fashion as the factorio wiki.
So if I got the calculations right, with a 0,02% chance to get one, I roughly need 23700-23800 cycles in order to get a 95% probability ?

nesklein
Manual Inserter
Manual Inserter
Posts: 1
Joined: Wed Apr 22, 2020 2:23 pm
Contact:

Re: Confidence level calculation

Post by nesklein »

DaveMcW wrote:
Mon Apr 20, 2020 7:48 pm
POISSON.DIST() Excel function

failure_chance = POISSON.DIST(required_number_of_successes, success_chance * number_of_events, TRUE)

failure_chance = POISSON.DIST(40, 0.007 * 7000, TRUE)

failure chance for 7000 cycles = 10.98%
It really helped. Thank you very much.

Post Reply

Return to “General discussion”