I came up with a simple, elegant, one-line solution to a problem.  This solution fixed 98% of cases it was meant to address.

The solution for the other 2% involved 40 lines of code and for loops nested 3 levels deep:

#!/usr/bin/ksh

SP="www.xxx.yyy.zzz"
SCRIPT=$(basename $0)
SOURCE=/var/adm/scripts/SnapView
DATA=$SOURCE/data

if [[ -z $1 ]]
then
  echo "###############################################"
  echo "## No SID specified, exiting."
  echo "## Example: $SCRIPT dev1 "
  echo "###############################################"
  exit 1
fi

SID=$1

OUTPUT_FILE=$DATA/${SID}_recreate.txt
HDISK_LIST=`powermt display dev=all | grep Pseudo | awk -F"=" '{print $2}'`

cp /dev/null $OUTPUT_FILE

for VG in `cat $DATA/${SID}_vg_to_lun_map.txt | awk '{print $1}'`
do
        HDISK=""
        for LUN in `grep $VG $DATA/${SID}_vg_to_lun_map.txt | awk '{$1=""}1'|sed 's/^ //'`
        do
                UID=`naviseccli -h $SP getlun $LUN | grep UID | tr -d ":" | awk '{print $2}'`
                        for x in $HDISK_LIST
                        do powermt display dev=$x | grep $UID > /dev/null && HDISK="$HDISK $x"
                        done
        done
        echo $VG $HDISK >> $OUTPUT_FILE
done

Impressive as always.  Depending on the output of your first awk statement might be modified to run on HP/UX, but since you're using volume groups (assuming AIX) that'll translate nicely.  Timely too, since Steve accidentally mapped the $18,000-per-drive SSD array last week and had to work double-quick to get everything moved before end-of-month disk reports came out and they charged us for those capacity-on-demand drives.

P.S.  Why doesn't it look like a Christmas tree?