Calculate task forecasts with time intelligence based on summing a value

[AspiringAnalyst]

Hello all,

My business case is that I have a nursery.  I'm recording GDUs (growing degree units) each day and have a calculated column set up that if I populate my known GDU accumulated by day it uses that value (good for past dates and today).  I forecast avg historical GDU into the calculation for future dates, and I have the planting dates for my crops.  How can I predict the task date based off of summing the GDUs until the known GDU requirement has been met?

Here is my planting list table of required GDU for task and the date it was planted:

'checklist'

Variety	GDU required	Planting Date
Bell Pepper A	1401	4/15/2024
Bell Pepper B	1360	4/21/2024
Bell Pepper C	1360	4/30/2024
Bell Pepper D	1314	5/3/2024
Bell Pepper E	1353	5/5/2024
Bell Pepper F	1290	5/8/2024
Cantelope A	1000	4/25/2024
Cantelope B	1060	4/30/2024
Watermelon A	1450	4/15/2024
Watermelon B	1400	4/21/2024
Watermelon C	1425	4/30/2024
Tomato A	1337	4/30/2024
Tomato B	1324	5/3/2024
Tomato C	1312	5/5/2024
Tomato D	1353	5/8/2024
Spinach A	650	5/3/2024
Spinach B	700	5/5/2024
Spinach C	630	5/8/2024

Here is my GDU table (I know it's long), where each day I input actual GDU, and future dates are the avg/day historically:

'GDU tracking'[CalculatedGDUs] being the column of interest

Date	Low	High	GDU	CalculatedGDUs
5/1/2024	52	71	11.5	11.5
5/2/2024	54	66	10	10
5/3/2024	42	73	11.5	11.5
5/4/2024	48	68	9	9
5/5/2024	40	69	9.5	9.5
5/6/2024	54	76	15	15
5/7/2024	57	77	17	17
5/8/2024	52	81	16.5	16.5
5/9/2024	54	65	9.5	9.5
5/10/2024	51	73	12	12
5/11/2024	46	78	14	14
5/12/2024	49	86	18	18
5/13/2024	61	73	17	17
5/14/2024			0	15
5/15/2024			0	15
5/16/2024			0	15
5/17/2024			0	15
5/18/2024			0	15
5/19/2024			0	15
5/20/2024			0	15
5/21/2024			0	15
5/22/2024			0	15
5/23/2024			0	15
5/24/2024			0	15
5/25/2024			0	15
5/26/2024			0	15
5/27/2024			0	15
5/28/2024			0	15
5/29/2024			0	15
5/30/2024			0	15
5/31/2024			0	15
6/1/2024			0	21
6/2/2024			0	21
6/3/2024			0	21
6/4/2024			0	21
6/5/2024			0	21
6/6/2024			0	21
6/7/2024			0	21
6/8/2024			0	21
6/9/2024			0	21
6/10/2024			0	21
6/11/2024			0	21
6/12/2024			0	21
6/13/2024			0	21
6/14/2024			0	21
6/15/2024			0	21
6/16/2024			0	21
6/17/2024			0	21
6/18/2024			0	21
6/19/2024			0	21
6/20/2024			0	21
6/21/2024			0	21
6/22/2024			0	21
6/23/2024			0	21
6/24/2024			0	21
6/25/2024			0	21
6/26/2024			0	21
6/27/2024			0	21
6/28/2024			0	21
6/29/2024			0	21
6/30/2024			0	21
7/1/2024			0	23
7/2/2024			0	23
7/3/2024			0	23
7/4/2024			0	23
7/5/2024			0	23
7/6/2024			0	23
7/7/2024			0	23
7/8/2024			0	23
7/9/2024			0	23
7/10/2024			0	23
7/11/2024			0	23
7/12/2024			0	23
7/13/2024			0	23
7/14/2024			0	23
7/15/2024			0	23
7/16/2024			0	23
7/17/2024			0	23
7/18/2024			0	23
7/19/2024			0	23
7/20/2024			0	23
7/21/2024			0	23
7/22/2024			0	23
7/23/2024			0	23
7/24/2024			0	23
7/25/2024			0	23
7/26/2024			0	23
7/27/2024			0	23
7/28/2024			0	23
7/29/2024			0	23
7/30/2024			0	23
7/31/2024			0

I'd like to forecast, based off of the planting date, adding GDUs in the 'GDU tracking'[CalculatedGDUs] column, the date on which I can expect my crop to flower, hit a milestone, etc.  

This would be a calculated column in my 'checklist' table named [ForecastedDate].  Ideally, I'm able to change the value of 'GDU required' in different columns based off of different task needs to generate a to-do list of tasks by date for each crop variety.

Where do I start with this?  I feel like I'm using VAR and the GDU required column in a CALCULATE function, though I'm not sure the right approach.

[lbendlin · Accepted solution]

[lbendlin · Accepted solution]

Please provide sample data that fully covers your issue.
Please show the expected outcome based on the sample data you provided.

[lbendlin]

should be possible. What's on the Y axis?

[AspiringAnalyst]

I have a measure set to "sum of number of plants" based off of my column of "number of rows" of that variety.  "Sum of number of plants" is set to "number of rows" * 20.  Categorized by crop in the legend.  I'm unable to set "Ready Date" to the x axis to show number of plants (workload) by day.  I have these dimensions set in the planted.pbix model you uploaded but the reply function isn't allowing me to upload it to show my example.  Not that I can figure out anyway.

[lbendlin]

If you are unsure how to upload data please refer to https://community.fabric.microsoft.com/t5/Community-Blog/How-to-provide-sample-data-in-the-Power-BI-...

[AspiringAnalyst]

Hi @lbendlin, I have uploaded my pbix example to a dropbox.  I've tried to use ChatGPT to help convert the results to a calculated column though the results don't match up with your measure outputs.  Those are correct.

My end goal is to summarize number of plants by ready date and in what field.  Fields aren't in my example file though and that's not a concern really as that's already available in my working file and is easily sliced to summarize a given date's workload.  

I can't add the ready date measure to a slicer, and I'm struggling to add the ready date measure to an x axis.  I'd like to use a clustered column chart to show workload over time, and for daily task setting a table that summarizes number of plants by date and their fields.  Variety isn't that important for that table, that's just the level of granularity that the GDU is set to.

Is a calculated column possible?

https://www.dropbox.com/scl/fi/hnyceeoe34vbh2jt3228c/planted.pbix?rlkey=5nwpajoyby3v64swxlbr8eh44&dl...

[lbendlin · Accepted solution]

Here's the Ready Date as a calculated column in the checklist table

Ready Dt = 
var a = filter('GDU tracking',[Date]>=[Planting Date])
var b = ADDCOLUMNS(a,"c",var d = [Date] return sumx(filter(a,[Date]<=d),[CalculatedGDUs]))
var c = filter(b,[c]<=[GDU required])
return maxx(topn(1,c,[Date],DESC),[Date])

[lbendlin · Accepted solution]

[AspiringAnalyst]

Is it possible to set the measure of ready date to the x axis of a bar chart?  I am now looking to aggregate a count of things ready by date via a chart over a table.

[AspiringAnalyst]

Wow thank you!

I've transferred the measures into my working model and I see that in instances of varieties not yet populated with a planting date, it is summing from the beginning of the CalculatedGDUs column.  What is the best way to not assign Ready Dates to those items?  Would it be in the DAX expression itself or through a Filter on the page/visuals removing null values for planting date?

[lbendlin · Accepted solution]

Please provide sample data that fully covers your issue.
Please show the expected outcome based on the sample data you provided.

[AspiringAnalyst]

The only thing different was that I had a few varieties that weren't yet planted.  I was just asking your opinion on how you'd handle that, but I just set a filter to not show things not yet planted.

Thank you for your help.