Data delivered to a printer to show what a document looks on paper will always be inferior to the data present in the original application. Word knows headings, sentences, reading order, possibly image descriptions, and so forth. Trying to deduce logical structure from display information is possible but really hard in the general case.
For a specific doc or set of docs, you might be able to notice, for example, that "Arial Bold 14pt" is always and only used for heading level 2, so detecting heading level 2 is easy. For all possible docs? Not so much.
PDF/A 1-a is for accessability... screen readers for the blind for example. Text descriptions of tables and images, reading order, logical structure.
PDF/A 1-b is for archival purposes. You have to embed all your fonts for example.
I have Word 2016. When I "Save As..." to PDF, there's an "Options" button. One of the check boxes in the resulting dialog is labeled "ISO19005-1 compilant (PDF/A)".
It doesn't specify which PDF/A 1 it supports, A, or B. B is pretty easy, and supporting it at the printer level is no problem. A... not so much. Saving out a file with that box checked shows... PDF/A-1A. Good for them.
What PDF printer are you using, and why aren't you using the built-in "Save as PDF" function that Word already has? – user1686 – 2019-12-08T20:20:52.350
this is what i am recommending, however, there is a user education piece that needs to go along with this as they will need to change the way they produce pdf files. – w00ki3 – 2019-12-10T06:13:34.887
So once again, which PDF printer are you using? They all have independent options... – user1686 – 2019-12-10T06:14:40.980
Check out CutePDF. Our testing dept used to use it. I think the free one can do this. Here is page showing how to do it and Here is the CutePDF website
– Señor CMasMas – 2019-12-11T22:02:02.013