Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowTry your skills in the Power BI Dataviz World Championship! Round one ends June 26. Join now
Hi,
I'm trying to create a new column [New Qty] using DAX which convert the [Diff Qty] to the max values of [Usage Qty] in the same row and extend the [New Qty] to the next row until the sum equals the [Diff Qty], based on Product ID. Could anyone help me on that? Thanks.
Solved! Go to Solution.
Hi, @Troyes
You can click ‘Edit Queries’ to enter into the Query Editor, go to ‘Add Column’ -> click ‘Index Column’ to add index column (start from 0).

Then, you can create a calculated column as follows:
New QTY =
VAR t =
FILTER (
'Table',
[Product] = EARLIER ( 'Table'[Product] )
&& [Index] < EARLIER ( 'Table'[Index] )
)
VAR totalDiff =
CALCULATE ( SUM ( 'Table'[DIFF QTY] ), t )+[DIFF QTY]
VAR minleft =
CALCULATE ( SUM ( 'Table'[Usage QTY] ), t )
RETURN
IF (
minleft > totalDiff,
0,
IF (
[DIFF QTY] <> 0,
[Usage QTY],
IF (
minleft = 0,
0,
IF ( minleft + [Usage QTY] > totalDiff, totalDiff - minleft, [Usage QTY] )
)
)
)
Result:

If I misunderstand the your thoughts, please inform me of your expected output. I am glad to solve the problem for you.
Best Regards,
Allan
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Hi, @Troyes
You can click ‘Edit Queries’ to enter into the Query Editor, go to ‘Add Column’ -> click ‘Index Column’ to add index column (start from 0).

Then, you can create a calculated column as follows:
New QTY =
VAR t =
FILTER (
'Table',
[Product] = EARLIER ( 'Table'[Product] )
&& [Index] < EARLIER ( 'Table'[Index] )
)
VAR totalDiff =
CALCULATE ( SUM ( 'Table'[DIFF QTY] ), t )+[DIFF QTY]
VAR minleft =
CALCULATE ( SUM ( 'Table'[Usage QTY] ), t )
RETURN
IF (
minleft > totalDiff,
0,
IF (
[DIFF QTY] <> 0,
[Usage QTY],
IF (
minleft = 0,
0,
IF ( minleft + [Usage QTY] > totalDiff, totalDiff - minleft, [Usage QTY] )
)
)
)
Result:

If I misunderstand the your thoughts, please inform me of your expected output. I am glad to solve the problem for you.
Best Regards,
Allan
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Don't miss out on Data Days, June 15 through August 7. Learn Fabric, Power BI, SQL, AI and more.
Check out the May 2026 Power BI update to learn about new features.
| User | Count |
|---|---|
| 22 | |
| 22 | |
| 21 | |
| 20 | |
| 12 |
| User | Count |
|---|---|
| 58 | |
| 55 | |
| 41 | |
| 36 | |
| 34 |