I use XML::LibXML to do my XML parsing (fastest I've found, very complete, very reliable, clean interface), and I use XPaths (via findnodes) to extract data from the resulting tree.

Example:

my @forms = $xpc->findnodes('/plm:PLMXML/plm:Form');

for my $form (@forms) {
    print( $form->getAttribute('subClass'), "\n" );
    ...
}
[download]

Example:

my ($form) = $xpc->findnodes(
    '/plm:PLMXML/plm:Form[subClass="Part Revision Master"]'
)
    or die("Can't find Part Revision Master form\n");

print($form->getAttribute('id'), "\n");  # id24
[download]

Update: Added examples.
Update: Fixed examples.