Logistics Forecast Distribution

[Anonymous]

Hi everyone!

Im trying to forcasting how our trucks are going to be busy in the next day.

The information that I have is the shipping plant (have many), the order and the stimating time of start time of the travel and when is going to finish.

I dont know yet which truck im gonna use for each order until next day, but I need to forecast a day before to see if im not over scheduling.

Plant	Order	Start Time	Finish Time
510	1009	6:29:00 a. m.	8:24:00 a. m.
510	1192	6:35:00 a. m.	8:28:00 a. m.
510	1192	6:50:00 a. m.	8:43:00 a. m.
510	1023	7:09:00 a. m.	8:21:00 a. m.
510	1009	7:29:00 a. m.	9:24:00 a. m.
510	1023	7:44:00 a. m.	8:56:00 a. m.
510	1091	7:50:00 a. m.	9:33:00 a. m.
510	1023	8:19:00 a. m.	9:31:00 a. m.
510	1009	8:29:00 a. m.	10:24:00 a. m.
510	1305	8:29:00 a. m.	9:54:00 a. m.
510	1322	8:29:00 a. m.	9:54:00 a. m.
510	1091	8:50:00 a. m.	10:33:00 a. m.
510	1023	8:54:00 a. m.	10:06:00 a. m.
510	1360	8:55:00 a. m.	10:40:00 a. m.

This is what i have when the trucks are assigned, so im expecting something similar but with the estimation. (is a custom timeline visual, but if you have a better visual is welcome too):

What i was thinking is that depending on the start time, write something like a generate series from 1 to X (the numer of truck available), but if i made that measure it just drop me an error:

"A table of multiple values was supplied where a single value was expected."

Any sugestions?

[lbendlin]

you can use table variables in your measure but the final result of the measure must be a scalar value (a table with a single row and a single column).

What should the measure compute?

[Anonymous]

Im expecting to get another column with the numer I set in the what if parameter, get to that number and repeat based on the start time... i mean if I set de number 6, i would get the next information:

1

2

3

4

5

6

1

2

3

4

5

6

1

...

[lbendlin · Accepted solution]

Let's assume you have index numbers in a column called [IDX] from 1 to n.  Your DAX would be something like

Val = 1+MOD([IDX]-1,6)

[Anonymous]

Yes, I add an index measure and then another with the info you gave me and it works. Thanks!