Optimization in Proc MIRP

[ashishdhakad]

Hi All , I want to know to what optimization technique proc MIRP is using in the backend , is it LPP or NLPP or any other technique . If any one have any leads please respond . Its little urgent to going forward in project .

Thank you 

[ashishdhakad]

Hi All , I want to know to what optimization technique proc MIRP is using in the backend , is it LPP or NLPP or any other technique . If any one have any leads please respond . Its little urgent to going forward in project .

Thank you 

[ashishdhakad]

Hi All , I want to know to what optimization technique proc MIRP is using in the backend , is it LPP or NLPP or any other technique . If any one have any leads please respond . Its little urgent to going forward in project .

Thank you 

[Kurt_Bremser]

Please do not make muktiple posts with the same question. Keep everything in one thread; if need arises, a Super User will move it to the correct community.

[sbxkoenk]

Hello,

The MIRP procedure is designed for (Multi-echelon) Inventory Replenishment Planning.
It calculates optimal inventory policies.

See here for the doc.

https://go.documentation.sas.com/doc/en/pgmsascdc/v_024/irpug/irpug_mirp_syntax02.htm?homeOnFail

I think what is not in the doc (all mathematical formula used in the procedure) are strictly company confidential. 

@JinxinYi : Some years ago, you shared with me some literature that was used as the basis for developing PROC (M)IRP.
I do not find these papers anymore (it was a mail with the titles and scientific magazines these papers were published in)!
Can you share these paper titles again?
Funny enough, the above link (from SAS DOC) has no "References" section.

Thanks,

Koen

[ashishdhakad]

@JinxinYi hey if you can please send it would be really helpful .

[ashishdhakad]

Thanks Koen for the information

[sbxkoenk]

Hello @ashishdhakad ,

I have sent @JinxinYi an internal mail (to his SAS inbox) with the request to respond here.

Thanks,

Koen

[Xinmin]

Proc mirp optimizes inventory policy over assembly or distribution networks, the core algorithm is based on news vendor type of problem (single location, single period, and stationary demand), but with extension to multiple periods with non-stationary demand as well as to a distribution or assembly network. On the top of that, it has a built-in simulator which handles the non-stationary demand pattern in the real world. It offers order up to level or (s,S) policy out of box. Jinxin and I have a U.S patent for proc mirp,

https://patents.google.com/patent/US8515835B2/en, there are some technical details in the doc as well. Hope it helps.

Xinmin

[ashishdhakad]

Thank you so much @Xinmin  , it will help.