Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi,
I'm trying to calculate the actual takt time per station.
Example: 1 Machine glues two parts together. Once it's done it safes a timestamp. This is done with every part.
However the machine produces different parts, based on settings. It doesn't know if there's a shiftbreak, a holiday etc.
Therefor what I'm trying to do is to calculate takt time " Timestamp - Previous(Timestamp) AS Taktzeit, "
And now I'd like to calculate the avg Taktzeit where Takzeit < 2*(Stdev(Taktzeit)).
This would eliminate all shift breaks + larger machine crashes etc. while still showing minor erros if timings go up or down.
I can't wrap my head around the most elegant solution for this though.
Any ideas?
Thank you!
If possible share some your data with us
Try with the formula:
AVG(IF(Taktzeit < 2/24, Taktzeit))
Guess that Taktzeit will be part of a day, therefore 2 hours are 2/24 days.
Peter
You're right, I'm sorry, here's some example data:
Load
Ident,
Timestamp,
$Measure,
Type,
Timestamp - PREVIOUS(Timestamp) AS Taktzeit,
Ident | Timestamp | $Measure | Type |
930982 | 42008,9048 | 31 | 1 |
930983 | 42008,9051 | 46 | 1 |
930984 | 42008,9053 | 46 | 1 |
930985 | 42008,9054 | 43 | 1 |
930986 | 42008,9057 | 46 | 1 |
930987 | 42008,9058 | 49 | 1 |
930988 | 42008,9095 | 184 | 1 |
930989 | 42008,9098 | 209 | 3 |
930990 | 42008,9099 | 215 | 3 |
930991 | 42008,9101 | 190 | 3 |
930992 | 42008,9103 | 209 | 4 |
930993 | 42008,9105 | 206 | 5 |
930994 | 42008,9217 | 193 | 5 |
930995 | 42008,9219 | 202 | 5 |
930996 | 42008,9221 | 190 | 4 |
930997 | 42008,9223 | 206 | 4 |
930998 | 42008,9225 | 196 | 4 |
930999 | 42008,9227 | 215 | 4 |
931000 | 42008,9229 | 193 | 2 |
931001 | 42008,9231 | 212 | 2 |
931002 | 42008,9233 | 212 | 2 |
931003 | 42008,9235 | 212 | 2 |
I haven't included Taktzeit since it ignores the Type atm.
=Time(avg(Taktzeit)) = Value including gabs between non production days + breaks. Atm. @ ~25 seconds.
my thought was to use a set analysis like this: "=time(avg({<$={'>0'}>+<$={'<00:00:35'}>}Taktzeit))"this doesn't give me the expected result though. It still shows the common avg which is way to high due to breaks and non production days
So all my solutions currently ignore the type thingy. However thinking about it it might be better to not calculate within the script but in the application itself to be able analyze including all types as well as single types by themselves as well.
Any ideas?
I think your last expression shows incorrect syntax in the set expression, try something like this instead:
=Interval(Avg({<Taktzeit= {"<=$(=Num('00:00:35','.',','))"} >}Taktzeit))