Ideal sequence of nightly MRP/Global Reschedule

Yahoo Archive
1
Robert,



Id be glad to discuss that with you, we spent a lot of time with our
scheduling system, and were fairly happy with the direction its going.
There are a lot of variables there. Id be happy to field a call
offline... I don't feel like typing a book right now J



Rob Bucek

Manufacturing Engineer

PH: (715) 284-5376 ext 3111

Mobile: (715)896-0590

FAX: (715)284-4084

<http://www.dsmfg.com/>

(Click the logo to view our site) <http://www.dsmfg.com/>





From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com] On Behalf
Of Robert Brown
Sent: Tuesday, May 11, 2010 12:32 PM
To: vantage@yahoogroups.com
Subject: [Vantage] Ideal sequence of nightly MRP/Global Reschedule





Looking for opinions:

We're on 8.03.405a progress w/ AMM & have been live since June '08.

Because of the order volume, product sales variation & velocity of our
business, we run full MRP regens nightly (scheduling enabled, PO
suggestions enabled & pegging enabled) & also run Calc Global Scheduling
Order/Global Scheduler nightly.

While doing pre-upgrade testing of 409a, we've realized we have
apparently been running the Calc-Sched-Order/Global-Sched pair process
1st followed by the full MRP regen.

Piloting was so long ago I no longer recall our reasoning (if there was
any?! ) & it strikes me that this may not be the best sequence (as any
unfirmed MRP jobs will find no near term capacity after the global
tightly repacks resource capacity with existing work - distorting
unfirmed job start/due dates and any resulting lower level purchased
material requirement dates that result in Change or New PO suggestions).

The argument I can think of FOR running global 1st and then MRP is that
it is more imperative that firmed (by a human planner) jobs & resulting
material requirement dates be globally rescheduled (& stable) so that
any resulting New/Change PO messaging is valid for them - and we have
several thousand firmed/engineered jobs at any given time versus between
25-125 MRP generated unfirmed jobs nightly. (I think I just answered my
own question? /;o )

What are others of you out there doing? MRP or Global 1st (if you do
both) - and why?

Any feedback appreciated.

Thanks

Rob Brown
Versa Products





[Non-text portions of this message have been removed]
2
Looking for opinions:

We're on 8.03.405a progress w/ AMM & have been live since June '08.

Because of the order volume, product sales variation & velocity of our business, we run full MRP regens nightly (scheduling enabled, PO suggestions enabled & pegging enabled) & also run Calc Global Scheduling Order/Global Scheduler nightly.

While doing pre-upgrade testing of 409a, we've realized we have apparently been running the Calc-Sched-Order/Global-Sched pair process 1st followed by the full MRP regen.

Piloting was so long ago I no longer recall our reasoning (if there was any?! ) & it strikes me that this may not be the best sequence (as any unfirmed MRP jobs will find no near term capacity after the global tightly repacks resource capacity with existing work - distorting unfirmed job start/due dates and any resulting lower level purchased material requirement dates that result in Change or New PO suggestions).

The argument I can think of FOR running global 1st and then MRP is that it is more imperative that firmed (by a human planner) jobs & resulting material requirement dates be globally rescheduled (& stable) so that any resulting New/Change PO messaging is valid for them - and we have several thousand firmed/engineered jobs at any given time versus between 25-125 MRP generated unfirmed jobs nightly. (I think I just answered my own question? /;o )

What are others of you out there doing? MRP or Global 1st (if you do both) - and why?

Any feedback appreciated.

Thanks

Rob Brown
Versa Products