Floating error

Use this forum to post Vensim related questions.
Post Reply
smaalmas
Junior Member
Posts: 3
Joined: Tue Jun 12, 2018 1:15 pm
Vensim version: Ventity

Floating error

Post by smaalmas »

Hello,

Please, I want to help me
I have a problem in the developed model the message is:
Floating point error computing - Escalation to cost - at time = 2009.625000.
Trying to save the results anyway
If any one could help me please,
If I have changed the time step there is no problem.
I want to ask practically, what is the meaning of time step ,I mean why the model runs ok if the time step changed and how I can solve the problem



Thanks
Administrator
Super Administrator
Posts: 4573
Joined: Wed Mar 05, 2003 3:10 am

Re: Floating error

Post by Administrator »

Are you using Vensim or Ventity?
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
tomfid
Administrator
Posts: 3806
Joined: Wed May 24, 2006 4:54 am

Re: Floating error

Post by tomfid »

Ventity creates NaN values, so this must be a Vensim question.
tomfid
Administrator
Posts: 3806
Joined: Wed May 24, 2006 4:54 am

Re: Floating error

Post by tomfid »

tomfid
Administrator
Posts: 3806
Joined: Wed May 24, 2006 4:54 am

Re: Floating error

Post by tomfid »

grayven87
Junior Member
Posts: 15
Joined: Wed Apr 17, 2019 3:54 pm
Vensim version: PLE

Re: Floating error

Post by grayven87 »

Hi everyone

my model is giving me this error for a flow variable of mine at time 0 whenever I try to run it.can anyone help me with this issue and please state:
-Is the issue with the variable equations?
-Should I be looking at the cause diagrams and fix that?
-Could it be related to the whole model(because the flow variable that has an error is the last flow variable in the model)?
-Is there anyway I can find the wrong part of the model in Vensim, like in programming languages (like when coding, there is an option that guesses if the logic could be wrong)?
Administrator
Super Administrator
Posts: 4573
Joined: Wed Mar 05, 2003 3:10 am

Re: Floating error

Post by Administrator »

Most floating point errors are due to division by zero. If you can upload your model, we can take a look.
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
tomfid
Administrator
Posts: 3806
Joined: Wed May 24, 2006 4:54 am

Re: Floating error

Post by tomfid »

Another common possibility is overflow from a runaway positive feedback loop or EXP(something big).

You should be looking at the variable that is reported as the failure at runtime. The table tool is often the easiest way to debug - set it to show causes, and you can easily see the problem variable and its inputs.
Post Reply