Specifying the type of feedback structure

Use this forum to discuss any issues relating to Systems Dynamics and Systems Thinking.
Post Reply
Monte
Senior Member
Posts: 133
Joined: Sat Aug 13, 2005 4:18 am
Vensim version: PLE+

Specifying the type of feedback structure

Post by Monte »

Dear Sirs,

When a table graph shows both positive and negative relationships between the input and output, how could we specify the polar (+/-) of the causual link that interlocks the multiplier and its multipliee (a normal constant) ? I think it would be hard to tell whether the feedback structure is of positive or negative one.

Thank you.
Monte
LAUJJL
Senior Member
Posts: 1427
Joined: Fri May 23, 2003 10:09 am
Vensim version: DSS

feeedback structure

Post by LAUJJL »

Hi

I use two ways to deal with that problem.

One way is just to leave things like they are, and draw a
loop with no sign in it, meaning that the loop can be positive or negative.

If one wants absolutely to know when the loop is negative or positive the solution is to replace the arrow that can be positive or negative by two arrows one always positive and one always negative if possible, and then you have two loops and arrows that are working one at a time.
Of course you have to find when the arrow is positive and when it is negative to be able to tell the model which arrow will be working.
So you can use an if then else like
Y = if then else (condition positive,F(x),G(X)) where F is the positive function and G the negative one.
Calculating the condition positive may be in certain cases complex.
Regards.
JJ
Monte
Senior Member
Posts: 133
Joined: Sat Aug 13, 2005 4:18 am
Vensim version: PLE+

Post by Monte »

This implies that sometimes a table graph showing both negative and positive relations can be judged valid.
Many thanks for your helpful note.

Monte
LAUJJL
Senior Member
Posts: 1427
Joined: Fri May 23, 2003 10:09 am
Vensim version: DSS

feedback structure

Post by LAUJJL »

Hi

You can have in the same lookup a positive and a negative influence at the same time.
Here is joined two models:
one with the loop with nothing in it with a negative and positive lookup, and one with two loops that gives the same result than the first but with the positive and negative loop splitted into two loops one negative and the other positive.
Regards.
Attachments
positive_negative.mdl
(2.59 KiB) Downloaded 329 times
LAUJJL
Senior Member
Posts: 1427
Joined: Fri May 23, 2003 10:09 am
Vensim version: DSS

feedback structure

Post by LAUJJL »

second model joined
Attachments
positive_and_negative.mdl
(3.34 KiB) Downloaded 318 times
LAUJJL
Senior Member
Posts: 1427
Joined: Fri May 23, 2003 10:09 am
Vensim version: DSS

feed back structure

Post by LAUJJL »

there was a mistake in the polarity of one of the arrow
joined the corrected model
Attachments
positive_and_negative.mdl
(3.34 KiB) Downloaded 318 times
Post Reply