• use strict and warnings
• Neaten your code with perltidy.
• Remove all code which has nothing to do with your problem: use List::Compare, etc.

Duly noted. Thanks for the tip on perltidy. I'd never heard of that script before.

What I am looking for is to remove the duplicate part numbers and combine the tags (and add the quantities) similar to below:

413: quantity=2.0000 ,description=PRESS GAUGE ,tags=PI-412 PI-400 ,par
+tnum=2204133000 ,
414: quantity=2.0000 ,description=THERMOWELL ,tags= ,partnum=220226100
+0 ,
415: quantity=5.0000 ,description=THERMOMETER ,tags= TI-100 TI-101 TI-
+200  ,partnum=2201176000 ,
[download]

The original information is below:

413: partnum=2204133000,description=PRESS GAUGE,quantity=1.0000,tags=P
+I-412 ,
414: partnum=2202261000,description=THERMOWELL,quantity=2.0000,tags= ,
415: partnum=2201176000,description=THERMOMETER,quantity=2.0000,tags= 
+,
576: partnum=2204133000,description=PRESS GAUGE,quantity=1.0000,tags=P
+I-400 ,
582: partnum=2201176000,description=THERMOMETER,quantity=3.0000,tags=T
+I-100 TI-101 TI-200
[download]

The code is below:

#!/usr/bin/perl
#
use strict;
use warnings;

my @tmp2 = (
"413: partnum=2204133000,description=PRESS GAUGE,quantity=1.0000,tags=
+PI-412",
    "414: partnum=2202261000,description=THERMOWELL,quantity=2.0000,ta
+gs=",
    "415: partnum=2201176000,description=THERMOMETER,quantity=2.0000,t
+ags=",
"576: partnum=2204133000,description=PRESS GAUGE,quantity=1.0000,tags=
+PI-400",
"582: partnum=2201176000,description=THERMOMETER,quantity=3.0000,tags=
+TI-100 TI-101 TI-200"
);

#
# Move the data into a hash %HoH
#

my $who   = ();
my $field = ();
my %HoH   = ();
my $rec   = ();
my $key   = ();
my $value = ();

foreach (@tmp2) {
    s/^(.*?):\s*//;
    $who       = $1;
    $rec       = {};
    $HoH{$who} = $rec;
    for $field ( split(/,/) ) {
        ( $key, $value ) = split /=/, $field;
        $rec->{$key} = $value;
    }
}

print "\nHash complete\n";

##############################################
##
## Remove duplicate part numbers and combine the tags, if necessary.
##
##

print "######################################";
print "Duplicate Removal Testing\n";
my %tmpHoH = %HoH;

my $keyouter;
my $valueouter;
my $keydup;
my $valuedup;

my @keydelete = ();

while ( ( $keyouter, $valueouter ) = each %HoH ) {
    while ( ( $keydup, $valuedup ) = each %tmpHoH ) {
        if ( $keyouter eq $keydup ) {
            next;
        }
        if ( $HoH{$keyouter}{partnum} eq $tmpHoH{$keydup}{partnum} ) {
            print "**********\n";
            print
"key: $keyouter, value: $HoH{$keyouter}{partnum}, tag= $HoH{$keyouter}
+{tags}\n";
            print
"key: $keydup, value: $tmpHoH{$keydup}{partnum}, tag= $tmpHoH{$keydup}
+{tags}\n";
            $HoH{$keyouter}{tags} =
              $HoH{$keyouter}{tags} . " " . $tmpHoH{$keydup}{tags};
            print "$tmpHoH{$keyouter}{tags}\n";
            print "**********\n";
            push( @keydelete, "$keyouter=>$keydup" );
        }
    }
}

print "######################################\n";

##############################################

#############################################
#
## Testing to see if the hash is working.
##  Uncomment Section below to test.

my $iter  = ();
my $iter2 = ();

for $iter ( keys %HoH ) {
    print "$iter: ";
    for $iter2 ( keys %{ $HoH{$iter} } ) {
        print "$iter2=$HoH{$iter}{$iter2} ,";
    }
    print "\n";
}
[download]

Here's how I'd do it:

use strict;
use Data::Dumper;

my $delimiter = "|";  # delimiter for multiple tags
my %data;
while(my $line = <DATA>) {
  chomp $line;
  $line =~ s/^(.*?):\s*//; # remove leading number and colon, and any 
+whitespace
  
  my $rec = {};
  my $partno;
  for my $field ( split(/,/, $line) ) {
    my ($key, $value) = split(/=/, $field);
    if ($key eq 'partnum') {
      $partno = $value;
    } else {
      $rec->{$key} = $value;
    }
  }

  
  if ( defined $data{ $partno } ) {
  
    if ( $data{$partno}{'tags'} ) {
      $data{$partno}{'tags'} .= $delimiter . $rec->{'tags'};
    } else {
      $data{$partno}{'tags'} = $rec->{'tags'};
    }
    
    $data{$partno}{'quantity'} += $rec->{'quantity'};
    
    unless ($data{$partno}{'description'} eq $rec->{'description'}) {
      warn "Multiple descriptions for $partno ! \n";
    }
    
  } else {
    $data{ $partno } = $rec;
  }
  
}

print Dumper(\%data), "\n";

__DATA__
413: partnum=2204133000,description=PRESS GAUGE,quantity=1.0000,tags=P
+I-412
414: partnum=2202261000,description=THERMOWELL,quantity=2.0000,tags=
415: partnum=2201176000,description=THERMOMETER,quantity=2.0000,tags=
581: partnum=2204227002,description=TEMP TRANSMITTER,quantity=1.0000,t
+ags=TE/TT-102
582: partnum=2201176000,description=THERMOMETER,quantity=3.0000,tags=T
+I-100 TI-101 TI-200
576: partnum=2204133000,description=PRESS GAUGE,quantity=1.0000,tags=P
+I-400
[download]

Notes:

• I use the part number as the main hash key.
• I use 'defined' to check for duplicates.
• If I find a duplicate, I append the tag, if another tag exists. (it would be better to keep the tags in an array, to avoid having to split this list later)
• I separate tags using the character in $delimiter, so you can make it whatever you like.
• If I find a duplicate, I combine the quantities, and sanity check the descriptions (should be the same?) - consider this optional.

I get this output:

$VAR1 = {
          '2201176000' => {
                            'quantity' => 5,
                            'description' => 'THERMOMETER',
                            'tags' => 'TI-100 TI-101 TI-200'
                          },
          '2204133000' => {
                            'quantity' => 2,
                            'description' => 'PRESS GAUGE',
                            'tags' => 'PI-412|PI-400'
                          },
          '2202261000' => {
                            'quantity' => '2.0000',
                            'description' => 'THERMOWELL',
                            'tags' => ''
                          },
          '2204227002' => {
                            'quantity' => '1.0000',
                            'description' => 'TEMP TRANSMITTER',
                            'tags' => 'TE/TT-102'
                          }
        };
[download]

Anyway, my main point is that it might be better to store your data by part number.

$HoH{$part_number}{quantity}= $qty;
$HoH{$part_number}{description} = $description;
...
[download]

I normally would store the information by part number however the file that produces this information is extremely large and has duplicate part numbers. This script is actually trying to combine and reduce the data to single part numbers as well as group the tags and add the quantities.

Storing the information by part number does what you are looking to accomplish, see the response of scorpio17 which provides an example of what I'm talking about.

As you are building the HoH, why not store partnums in a temporary hash; and not add to the HoH any record whose partnum has already been seen?