R.G Coyle (Practical Approach to SD) - Simultaneous Equation

karankhosla · Post by **karankhosla** » Tue May 19, 2009 9:58 am

Dear Jean-Jacques, if you are online, this question is for you!

Why does the author create a rate (PR.KL) which is dependent on it's own average (APR.K) thus creating set of simultaneous equations? This requirement is of the policy control part of the DMC model (Page 217 of R.G Coyle's book "SD a practical approach").

The model simply will not run in VENSIM with this policy modification; is there a reason why Prof Coyle does not pick up on this and what can I do to rectify the problem?

Model Attached.

Best

Karan

Wilson · Post by **Wilson** » Tue May 19, 2009 12:13 pm

Karan

I don't think its the basing of "production rate" on "average production rate" that is the issue. That is often used to create a stable production rate with additional fluctuation in demand.

It's a very long time since I've looked at Cosmic models. Fortunately I found a floppy disk driver to load the disk that came with Geoffs book.

The reason your model has a simultaneous eqn error is due to the intialisation value of the Raw Material Stock (RMS).

If the DMS model you are referring to is that in Fig 7.8 of the book, the intial value of RMS is Desired Raw Material Stock (DMRS), not:

(PR production rate-APR average production rate)*TARMS+WAP raw material stock coverage*APR average production rate

as you have it at the moment. If you replace the RMS initial value with DRMS you model runs okay, though I have not checked to see if the results match those in the book.

LAUJJL · Post by **LAUJJL** » Tue May 19, 2009 2:15 pm

Hi Karan

I join two models, slightly different from one another of the DMC problem, that work both. One is the simple model with mass balance and the second is with structural parameters to be optimized.
I do not remember having noticed any error in the DMC models but it is possible. Both models work.
I do not have anymore in my head the DMC problem, but if you have any problem I can send you other DMC models that correspond to different part of the books. I have about 10 more or less different DMC models.
Best regards.
JJ

[Edited on 19-5-2009 by LAUJJL]

karankhosla · Post by **karankhosla** » Tue May 19, 2009 4:59 pm

Dear Wilson

Wilson, indeed it does work!

However, I don't quite understand why I have to use DRMS as the initial value for RMS instead of (PR production rate-APR average production rate)*TARMS+WAP raw material stock coverage*APR average production rate

Shouldn't equilibrium initializations work the same wether you input a number or a formula?

Thank you for your time!

Karan

karankhosla · Post by **karankhosla** » Tue May 19, 2009 5:04 pm

Dear JJ,

Thank you! If you have a few moments, could you please email me what you think might be useful (to use as a reference, of course!):

karankhosla@gmail.com

By the way, good advice on the book!

Best

Karan

LAUJJL · Post by **LAUJJL** » Wed May 20, 2009 12:13 pm

Hi Karan

I do not understand what you mean by 'might be useful'.
Useful to what? Can you precise your thought.
Regards.
JJ

karankhosla · Post by **karankhosla** » Wed May 20, 2009 12:43 pm

Hi JJ,

I was just taking you up on your offer to share the other models you might have built for the DMC problem in Coyle's book.

Best.

Karan

[Edited on 5-20-2009 by karankhosla]

LAUJJL · Post by **LAUJJL** » Wed May 20, 2009 1:23 pm

Hi Karan

I do not know what is useful for you.

I think that to really understand Coyle's book, it is better to work with a different software than with the original Cosmic software, as it obliges you to translate the model into another language. Giving you the answer might not be the best thing for you. But if you really want me to send you the
models I have made, it is up to you.
It is the same with the Vensim user guide. I think that one has really understood a chapter when one is able to rebuild with no help all the models built in the chapter. If one is not able one has to start studying the chapter again.
Regards.
JJ

karankhosla · Post by **karankhosla** » Wed May 20, 2009 2:00 pm

Hi JJ,

Thank you.... I have, in fact, worked out the DMC problem in Vensim and am not looking for a solution; however, I am curious, about seeing if my results match up with your output and how our practical modeling techniques differ - I think this is a good way to learn from a seasoned SD'r as yourself!

Not to worry though, I will let you know if I have any further issues!

Best Wishes,

Karan

[Edited on 5-20-2009 by karankhosla]

[Edited on 5-20-2009 by karankhosla]

LAUJJL · Post by **LAUJJL** » Wed May 20, 2009 2:26 pm

Hi Karan

Here is joined everything I have found about things that I have stored about the DMC problem. Some models are just diagram and have no equations. I have joined some VPD and VOC files for optimization.
You say that I am a seasoned SD'er.
There is a saying in French that says:
'Au royaume des aveugles les borgnes sont les rois'
In the kingdom of blinds the one-eyed is the king'.
Of course I do not take into account the Forum administrators.
Regards.
JJ

[Edited on 20-5-2009 by LAUJJL]

karankhosla · Post by **karankhosla** » Wed May 20, 2009 4:01 pm

JJ - much appreciated!