Hi All,
I am looking for a way to predict the probability of occurrence based on historical events.
Can anyone help?
Thanks
Hi All,
I am looking for a way to predict the probability of occurrence based on historical events.
Can anyone help?
Thanks
Last edited by Bliznaca; 01-15-2021 at 04:17 AM.
Not much to go on. The usual method I would expect is to look at the historical data, count/measure how many times the event occurred, count measure how many total somethings there were over the same time period. The estimated probability of occurrence is then the (count of occurrence)/(total something).
In your sample, you call them "defects". I don't see anywhere where you show the total or the total non-defect (though I assume you have that data somewhere). If you are basing your probability on both years' data, then perhaps the total defects would be =SUM(C2:D13). A similar SUM() function for the total production (assuming these are items produced). Then your estimate for the probability could be =SUM(C2:D13)/SUM(total production).
Does that help?
Originally Posted by shg
Hello MrShortly,
I added in the sample sheet 2 new columns. Total 2020 x defects and Total devices.
For example, in Jan 2020 there were 66 defects on 66 different devices.
The total devices are the number of all working devices at different customers.
Thank you for the explanation.
The
Using MrShorty's suggestion applied to the file in post #3.
If you want the overall probability of defect for 2020 then use the following in cell I2: =E2/F2
If you want a monthly probability* then use: =D2/(F$2/12) and drag the fill handle down to cell I13
As there is also data for 2019 it may be that you want a monthly probability* based on an average of 2019 and 2020 the defects for each month. If that is the case use: =AVERAGE(C2:D2)/(F$2/12)
*Note that the monthly probabilities assume that there are an equal number of devices produced each month.
Let us know if you have any questions.
Consider taking the time to add to the reputation of everybody that has taken the time to respond to your query.
There are currently 1 users browsing this thread. (0 members and 1 guests)
Bookmarks