Hi I am trying to create a table for Mean Time Before Failure (MTBF) over a 12 month rolling window. My Problem is my faults end Dec 20 and after that i need to keep the window moving forward by subtracting data from faults per months before 12 month window start and prior to Dec 20 I need to have cumulative sum of faults. Units represent the sum and subtraction before Dec 20.
Operating Hour | Months | Faults | Units | MTBF |
175200 | Aug-19 | 1 | 1 | 175200 |
175200 | Sep-19 | 1 | 2 | 87600 |
175200 | Oct-19 | 1 | 3 | 58400 |
175200 | Nov-19 | 1 | 4 | 43800 |
175200 | May-20 | 1 | 5 | 35040 |
175200 | Jul-20 | 1 | 6 | 29200 |
175200 | Aug-20 | 1 | 7 | 25029 |
175200 | Sep-20 | 1 | 8 | 21900 |
175200 | Oct-20 | 1 | 9 | 19467 |
175200 | Nov-20 | 1 | 10 | 17520 |
175200 | Dec-20 | 1 | 11 | 15927 |
175200 | Jan-21 | 0 | 7 | 25029 |
175200 | Feb-21 | 0 | 6 | 29200 |
175200 | Mar-21 | 0 | 5 | 35040 |
175200 | Apr-21 | 0 | 4 | 43800 |
175200 | May-21 | 0 | 3 | 58400 |
175200 | Jun-21 | 0 | 2 | 87600 |
Solved! Go to Solution.
Hi @ahsansalik ,
I don't think I'm following.
Could you post your expected outcome as well as the starting data? That would help us a lot.
M.
@mceleavey Hi,
sorry i wasn't very clear with my query. i actually just want the focus on the middle 3 columns where the Faults are the happening each month and Units is a running sum of the faults until Dec 20. Now after Dec 20 there arent any faults so i am actually looking at a sliding 12 month window for my calculation of MTBF. So before Dec 20 all the faults are adding up so units are running sum of the faults. After Dec 20 no fault happened but since its a 12 month sliding window so Jan21 will be looking back upto Jan 20 so will not include any faults from 2019. As the months increase so does the sliding window is moving forward so Feb 21 will be upto Feb 20 and will not include faults before it. This is the data extract from a data source.
MTBF is = Operating Hour/ Units
Operating Hour | Months | Faults | Units | MTBF |
175200 | Aug-19 | 1 | 1 | 175200 |
175200 | Sep-19 | 1 | 2 | 87600 |
175200 | Oct-19 | 1 | 3 | 58400 |
175200 | Nov-19 | 1 | 4 | 43800 |
175200 | May-20 | 1 | 5 | 35040 |
175200 | Jul-20 | 1 | 6 | 29200 |
175200 | Aug-20 | 1 | 7 | 25029 |
175200 | Sep-20 | 1 | 8 | 21900 |
175200 | Oct-20 | 1 | 9 | 19467 |
175200 | Nov-20 | 1 | 10 | 17520 |
175200 | Dec-20 | 1 | 11 | 15927 |
175200 | Jan-21 | 0 | 7 | 25029 |
175200 | Feb-21 | 0 | 6 | 29200 |
175200 | Mar-21 | 0 | 5 | 35040 |
175200 | Apr-21 | 0 | 4 | 43800 |
175200 | May-21 | 0 | 3 | 58400 |
175200 | Jun-21 | 0 | 2 | 87600 |
Are you saying you only have the first three columns and you would like the Units and MTBF columns to be calculated dynamically in Alteryx?
M.
yes that's correct
I was thinking of using generate rows tool for data after dec 20 as that where data ends.
Hey @ahsansalik
I had a go at this, but got different answers. Why were you expecting Feb-21 to have 6 Units rather than the 7 faults which occurred since Feb-20?
(I've attached the workflow anyway, in case it helps)