You could try assigning the intended name to "$outfile" from the beginning:

use strict;

my $infile = 'c:/RZ_Q.008';  # is it the same input file every time?
[download]

my $outfile = sprintf("%d.txt", time())

my ( $yr, $mo, $dy ) = (localtime)[5,4,3];
my $outfile = sprintf( "%04d%02d%02d.txt",$yr+1900,$mo+1,$dy );

# (I tend to format date-as-filename so that it will be
# easy to sort file names chronologically:  YYYYMMDD)
# you can add hours and minutes, if you like -- check
# `perldoc -f localtime`...
[download]

Now, the only issue is to control the scheduling for when the script gets executed (which is not a perl question, per se, unless you search for a "perl version of cron for ms-windows", or some such... but isn't there a scheduler utility that comes with ms-windows?)

Apart from that, the compulsive golfer in me sees a few places where you could simplify/condense the code a bit. I should restrain myself -- maybe you're just showing a subset of the whole application, and premature optimizations are a sin anyway, of course. But I can't resist...

# (assign $infile, $outfile as above, open IN, OUT as in your post, th
+en:

while (<IN>) {
    print OUT unless ( /774(0[346789]|1[123468])/ );
}
# but if the set of numerics to exclude changes regularly,
# you'll want to figure out how to convey the proper set
# to the script on start-up, and construct a regex along the
# lines shown in your original code.
[download]