Hello all,
I am interested in doing system dynamics model building. However, the simulation options are rather limited in free versions of system dynamics software (Vensim). Specifically, I would like to run my model with a huge number of parameter configurations. I believe Monte Carlo simulation is the way forward in such a situation.
My questions are as follows:
1. Is there a more versatile, free SD software than Vensim PLE available? If not (as I suppose is the case),
2. Is there a fundamental difference between running the simulations with a dedicated SD software or something else? For instance, I have SAS which provides Monte Carlo simulation possibilities for ODE systems which Vensim PLE does not provide (PROC MODEL in SAS). SAS also provides data-to-model fitting possibilities which the free version of Vensim does not provide.
Surely, SAS is not free either but I happen to have a lisence at my disposal...
Thanks!
System dynamics simulation with non-SD software
Re: System dynamics simulation with non-SD software
Update: I guess the answer to question 1 is no.
However, I'm still interested in your opinions on using non-SD software to simulate SD models. I.e. writing the same equations in another program and running that (e.g. SAS). Are there any pitfalls in doing that (in addition to the lack of intuitive GUI)?
Thanks!
However, I'm still interested in your opinions on using non-SD software to simulate SD models. I.e. writing the same equations in another program and running that (e.g. SAS). Are there any pitfalls in doing that (in addition to the lack of intuitive GUI)?
Thanks!
-
- Super Administrator
- Posts: 4590
- Joined: Wed Mar 05, 2003 3:10 am
Re: System dynamics simulation with non-SD software
Most users on here will be using Vensim (so the question of "any pitfalls in using non SD software" will not be one they face).
If the results are correct and you know your model is dimensionally correct, I cannot see any reason why you shouldn't use something else. It's straightforward to replicate simple models in Excel and many people use that for modelling.
Tony.
If the results are correct and you know your model is dimensionally correct, I cannot see any reason why you shouldn't use something else. It's straightforward to replicate simple models in Excel and many people use that for modelling.
Tony.
Advice to posters seeking help (it really helps us to help you)
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Units are important!
http://www.bbc.co.uk/news/magazine-27509559
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Units are important!
http://www.bbc.co.uk/news/magazine-27509559
Re: System dynamics simulation with non-SD software
A good strategy is often to prototype your model in Vensim, to get units balanced, behavior as desired, structure robust in extremes, etc. Then translating the system of equations to another language is fairly easy, and you have more confidence in the structure.
/*
Advice to posters (it really helps us to help you)
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Blog: http://blog.metasd.com
Model library: http://models.metasd.com
Bookmarks: http://delicious.com/tomfid/SystemDynamics
*/
Advice to posters (it really helps us to help you)
http://www.ventanasystems.co.uk/forum/v ... f=2&t=4391
Blog: http://blog.metasd.com
Model library: http://models.metasd.com
Bookmarks: http://delicious.com/tomfid/SystemDynamics
*/
Re: System dynamics simulation with non-SD software
Hello,
Thank you for the comments, Administrator and tomfid. Your suggestion, tomfid, makes good sense; Vensim is indeed good for building and testing model structure - superior to many other programs. I will do as you suggest.
Thanks again!
Thank you for the comments, Administrator and tomfid. Your suggestion, tomfid, makes good sense; Vensim is indeed good for building and testing model structure - superior to many other programs. I will do as you suggest.
Thanks again!