Your code does not compile. Please do yourself a huge favor and  use warnings; and  use strict; at the start of every file. This is especially important for a novice Perler.

Even with strict and warnings, your code still has syntactic problems. The monks like to see working code, even if it does not do all (or even most) of what you want. Please see Short, Self-Contained, Correct Example.

Some syntactic thoughts:

• $b[$j] =s~/"//g; What is this statement supposed to do? Will it even compile?
• if ($j==1 && my @arr = grep { $_ =~ /\"1"$/ } @b); Will this compile?
• else Will this compile?

Some best-practices thoughts:

• my @a = <$fh1>; Is this supposed to read all lines from the file handle? Then why not name the variable  @lines or  @input_lines (since you want to transform them and then produce something like an  @output_lines array) instead?
• Similar objections apply to array names like  @arr and  @b
• if ($j==1 && my @arr = grep { $_ =~ /\"1"$/ } @b); Assuming this actually compiled, which  @arr in the script is referred to in the following  @arr = map{$_ = "P"} @arr; statement? (There are two  @arr arrays floating around in the script.)
• @arr = map {$b[$j] - 1} (1..$#b); A similar  @arr confusion attaches to this statement ...
• print @arr; ... and to this.

These comments are dashed off quickly. Although it may seem that way, I don't mean to be blunt or dismissive. I hope you will take them as food for thought.

Thanks for pointing out all the un-proper defined in my code. I will follow your recommendation to improve it.

Here's a template for using Text::CSV that you can edit based on your needs (you should also install Text::CSV_XS). Note that $row is an arrayref, how to work with it is discussed in perlreftut.

#!/usr/bin/env perl
use warnings;
use strict;
use Text::CSV;

my $csv = Text::CSV->new({ binary=>1, auto_diag=>2,
    eol=>$/, always_quote=>1 });
open my $ifh, '<', 'NRoomch.txt' or die $!;
open my $ofh, '>', 'output.txt' or die $!;
while ( my $row = $csv->getline($ifh) ) {
    
    $row->[1]='P'; # your logic to modify 2nd column here
    
    $csv->print($ofh, $row);
}
$csv->eof or $csv->error_diag;
close $ifh;
close $ofh;
[download]

thanks for pointing me to the perlreftut. I love it !

I follow your method using HTML Table to print output.html but I alway get it printing to screen can you correct for me please?. I have worked out the value for the array. It's worth to spend weekend to learn the perlref

 
#!/usr/bin/perl
use strict;
use warnings;
my @values = ();
use HTML::Table;
my $table = HTML::Table->new(
    -cols        =>    3,
    -border        =>    1,
    -padding    =>    1,
    -head        => [('Header1','Header2','Header3')],
    );

my $sourceFile = "<C:/Perl-Script/File1.txt";

open my $ifh,$sourceFile or die $!;
open my $ofh, '>', 'C:/Perl-Script/output.html' or die $!;

while (my $row = <$ifh>)
{
    chomp($row);
    $row =~ s/"//g;
    @values = split(',',$row);
    $table ->addRow(@values);
}

$table->print ($ofh,@values);
close $ifh;
close $ofh;
[download]

Nirvana is Now or Never

$table->print ($ofh,@values); print
Prints HTML representation of the table to STDOUT
http://search.cpan.org/~ajpeacock/HTML-Table-2.08a/lib/HTML/Table.pm

print $ofh $table->getTable;
[download]

getTable
Returns a string containing the HTML representation of the table. Ibid.