Posted: Mon Jun 09, 2008 7:38 pm Post subject: Re: Does anybody know if there are any concrete simulations?
Greetings. I just joined the network this evening. My day job is a system dynamics modeler (the same field as Forrester); my evening job is worrying about my children's future. I enjoy the daytime more!
Anyway, I know this is an old, perhaps dead post, but I couldn't help notice you mentioned the world model. You might not have known, but there was an update to the book in 2004 with recent data, etc... You can see a link to the book here:
Also, there are versions of the WORLD3 model in Vensim or STELLA/iThink. I don't have the Vensim version, but can pass along the STELLA version for anyone interested.
Joined: Feb 12, 2006 Posts: 98 Location: New York State, USA
Posted: Wed Jun 11, 2008 6:39 pm Post subject: Re: Does anybody know if there are any concrete simulations?
Hi Chris,
I am interested in system dynamics and modeling & applying these to resource depletion etc.
I can't say I am any big expert exactly in "system dynamics" but I have a Masters degree in physics, so I have used differential equations to model the dynamics of physical systems, plus I've been developing software for coming up on forty years, so I have spent a lot of time both thinking of software as a system whose dynamics I want to understand, and using software as a tool to model aspects of the world, especially electronic microcircuits as it has turned out, and using the models to explore the dynamic behavior of those aspects of the world. Kind of a long sentence there, sorry!
WRT petroleum, one of my projects is to help people think of the logistic equation as a differential equation together with its family of solutions, rather than just as a "bell curve" that just happens to fit.
The next step is then to look at variations on the logistic differential equation.
One big question with any model is: where to draw the boundaries - what is outside the system, what is inside the system. The other big question is of course how detailed to make the models of the system's internals.
The world models of Forrester et al. clearly drew the boundaries at a very interesting place. As I understand the models, they tend not to be terribly detailed. It's their comprehensiveness that is special.
What is really cool about the logistic equation is that it is so simple.
So what I am trying to explore is how to move from the logistic equation in the direction of the world model, but just taking one or two simple steps. Of course the result will not be as comprehensive as the world model. But because of the simplicity, it might be easier to understand. Also it can focus debate on fewer factors. One problem with a world model is that it can be hard for people to agree about - there is enough in there, everybody can find something to disagree with!
It's like focusing on one article in an encyclopedia versus a whole encyclopedia.
I am not too good with interactive graphics and web stuff. I started programming in 1969! I think I still can program the drum card for an IBM 029 card punch! I made money in school drawing graphs on a 30 inch CalComp drum plotter! So I am a bit of an antique!
One of my fantasies is to build some code, maybe Java, to let people play with parameters and initial conditions on some simple extensions of the logistic differential equation. It can be a fun yet powerful way to learn about the behavior of a system.
How much money and expertise are required to run a model in STELLA/iThink?
I have never played with MatLab. Someday maybe I will register for some college class just to be able to get a student license. That is not cheap software! Plus there are all these (expensive) add-on components that one generally needs.
Posted: Wed Jun 11, 2008 7:11 pm Post subject: Re: Does anybody know if there are any concrete simulations?
A group called the Millenium Institute has developed a model of U.S. Energy, called "Threshold 21".
Quote:
This paper presents a significantly updated version of the T21 USA model that is designed to analyze
the major energy challenges and choices faced by the USA in the broader context of their relation to
the economy, social factors, and the environment, and with links to the rest of the world as necessary.
Analyzing these issues in a single, integrated framework is complex, but the T21 USA model builds on
the available expertise and detailed models in many sectors to achieve a highly transparent way of
presenting and visualizing the results of different assumptions and policy options, to promote
productive dialogue among different stakeholders, and to help design reasonable policies while
offering options to mitigate possible negative effects.
T21 does not present the ultimate, optimal solution. Rather, it allows users to examine the long term
plausible outcomes of different scenario options and assumptions to assist policy makers,
stakeholders, and the interested public make more informed decisions and reach agreements on how
to move ahead, instead of staying locked in disagreement and inaction.
I downloaded the T21 model and ran the simulations. The model includes projected oil and natural gas depletion, and allows you to change certain parameters such as recovery percentages for oil reserves, growth rate of renewable energy sources and so on. After running a simulation you can examine projections for population growth, food production or oil prices.
I was interested in simulating a full-blown "MonteQuest Overshoot and Die-Off Scenario." After setting the renewable energy sources and recovery rates to the minimum values, the simulation showed oil prices at $300/barrel and gas at $10/gallon by 2014 with food production sharply dropping off around 2035.
However, projections for economic and population growth were not affected even with the sharp drop off in food production and increase in oil prices. This suggests to me that the goals of the model were merely to show "it can be done", where alternative energy sources replace oil depletion without much economic impact.
I actually found it a little unsettling to discover that they did not include relationships between food production and population growth or oil prices and economic growth. With oil at $300/barrel one might expect to see some type of economic contraction, but instead the growth rate was just a bit smaller. There wasn't much variance between their best case and worst case scenarios.
Nonetheless, this model seems fairly comprehensive and includes many variables for economic and population data.
Maybe someone else can try it and see if they get different results.
All times are GMT - 6 Hours Goto page Previous1, 2
Page 2 of 2
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
News
Discussions
Resources
Members
Forum FAQ
Search
Usergroups
Profile
Log in to check your private messages
Log in
