I like your code++.

A couple of minor nits:
I personally try to avoid using subscripts in favor of assigning a name to variables in a split.
In this case, I'm not sure what the number represents (or the name), I'm sure the OP knows better a better description than us.
I would use a push instead of array assignment to @ordered, just because it seems more natural to me.
This is minor stuff - no problem at all with your code.

Update: I guess I don't know what is supposed to happen if say YOUSLAV appeared twice for 20055111 or whether that is even possible to occur. If that is possible, the OP should clarify.

#! perl -slw
use strict;
use Data::Dump qw[ pp ];

my( @ordered, %grouped );

while( <DATA> ) {
    chomp;
    my ($number, $name) = split ',', $_;
    push (@ordered, $number ) unless exists $grouped{$number };
    push @{ $grouped{ $number } }, $name;
}
#pp \@ordered, \%grouped;

print "$_|", join ',', @{ $grouped{ $_ } } for @ordered;

=prints
20055111|YOUSLAV,YURT,TENWIMPL
20011271|YOUSLAV,WUMARTHE
20011541|YOUSLAV,TENWIMPL
20102741|WEDLOFOU,YOUSLAV,YURT,KUPLYSO,TENWIMPL
20155505|YOUSLAV,YURT,TENWIMPL
20147155|YOUSLAV,KUPLYSO,FRIMA
=cut

__DATA__
20055111,YOUSLAV,
20055111,YURT,
20055111,TENWIMPL,
20011271,YOUSLAV,
20011271,WUMARTHE
20011541,YOUSLAV,
20011541,TENWIMPL,
20102741,WEDLOFOU,
20102741,YOUSLAV,
20102741,YURT,
20102741,KUPLYSO,
20102741,TENWIMPL,
20155505,YOUSLAV,
20155505,YURT,
20155505,TENWIMPL,
20147155,YOUSLAV,
20147155,KUPLYSO,
20147155,FRIMA,
[download]