West Desk Flow Summary Comments

Yan,

Based on my review conducted the morning of February 6th, here is a list of
items that need to be addressed in the next version of the West Desk's Flow
Summary sheet:

PGT - These numbers appear to be one day off. The numbers in the Test Sheet
for February 6th are identical to those used by the West Desk yesterday for
February 5th. I verified that they are the Cycle 2 numbers posted on PGT's
website for the 5th. Generally, we use the Cycle 2 numbers, however,
sometimes only Cycle one is available for our morning meeting.

Northwest - None of these numbers match our flow sheet. It appears the Test
Sheet pulls Cycle 3 nominations, whereas we use Cycle 6. However, the
website we use to pull the flows is generally not available to use until as
late as 7 am (we are not sure why). I understand you may pull the numbers
from a different place, so we should find out whether you can get the numbers
earlier.

TW - TW numbers appear to be correct, except for the IB Link number. We pull
the IB link number from the El Paso Blanco I/C point reported by TW. It
looks like your sheet pulls the number from the El Paso INWPLBLA point. They
differ by about 5,000, but I don't know if one is more correct than the
other. To be consistent, though, we should probably change the Test Sheet.

PGE - The numbers also appear to be one day off. The numbers for February
6th are identical to our Feb. 5th numbers. Is this a reporting or a
scrubbing problem? It is possible you are pulling the data from the website
before they update it with the next day's data. We manually type these
numbers in. If you pull from our spreadsheet, then the latest numbers would
not be in when this report is run.

El Paso - Most of the numbers appear to be correct, except for the plant
receipts. The numbers are close, but we should compare our formulas because
I don't think this is a cycle problem.

General - The "Difference" column of the Test Sheet compares the latest cycle
for the current with the final cycle of the previous day. This can cause
intraday fluxuations in scheduled quantities to appear as day to day
fluxuations. Any "Difference" column should compare identical cycles. I
suggest that we add additional cycles to the Test Sheet for the current day
and the previous day. This way we can compare identical cycles as well as
pick up any intraday fluxuations.

As we discussed, you'll let me know when you're ready to revise this report.
These notes should give us a great place to start. I look forward to getting
this report ready to use.

Mat