powershell script vs for loop w/IPA nodes

I need some advice with a new IPA process
We have Lawson S3 v10 and IPA v10.1.1.58

I am creating an IPA that automatically update the PA16
The file I'm receiving can not be sorted or changed before I get it. The records are random, meaning there are multiple PA16's to be updated for the same person, but the records are NOT grouped together they can be 3rd record, 12th record and 22nd record.

I want to update Each Employee's PA16 screen once. So I need to order the file or loop through all of the records to group the ones I want. I also need to check if the pa16 record already exists and if so pull the seq number.

I have a different IPA that updates the PA16 for a specific system. Even though I only want to update specific codes, the Lawson Transaction node brings back all of the PA16 codes for one screen. Meaning the field_name_r1 ...r2...are previously populated with what exists now and they are NOT in seq number order. It appears they display in Name order.

So my question boils down to: what is the best way to do this? To get the files in the correct order? and/or is there a way to populate the pa16 (transaction node) with out bringing in all the existing data? I've gone back and forth about writing a powershell scrip to re-order the file before IPA reads it or having IPA read in all the records, loop through and pull the recs I need.

What would you do? and any other words of wisdom you'd like to share?