Similar words are combining into one entry in the index

Weird behavior! …

What you need is a Script that creates your index! 1 script, 1 click!

Let's take your example:

Instead of indexing "RSR-0003", the Script will index "RSR-aaac" (a = 0, c = 3).

Instead of indexing "RSR-003", the Script will index "RSR-aac" (a = 0, c = 3).

[Of course, the Script will do this treatment itself!]

After having generated the final index:

RSR-aaac

RSR-aac

… the Script will replace in its text:

"a" by "0"

"b" by "1"

"c" by "2"

"d" by "3"

"e" by "4"

"f" by "5"

"g" by "6"

"h" by "7"

"i" by "8"

"j" by "9".

So, in your sample, the index will finally be:

RSR-0003

RSR-003

[correctly sorted]

(^/)  The Jedi

Ok - I think what is happening there is that InDesign thinks both entries are the same because it ignores punctuation when sorting and matching index terms. It treats RSR0003 and RSR003 as identical. Great for novels, absolutely terrible for SKU lists.

I would imagine in my head this how it might work to fix

1. Force InDesign to treat them as different terms
Tell it to sort by something else. Open the New Index Entry dialog and use the Sort By field.
For example:
Topic: RSR 0003
Sort By: RSR 0003

Then for the next one:
Topic: RSR 003
Sort By: RSR 003

That extra control might stop InDesign from flattening them into one entry.

2. Use a hidden character to differentiate them
If you want the visible text to remain exactly the same, you can insert a thin space or a zero width non joiner in the Sort By field. It still sorts separately but displays cleanly.

It is a bit fiddly, but it might work.

Might be that SKU indexing is one of the places where InDesign is a bit too clever for its own good.