|
|
|
|
Subject: Re: agent-based modelling and simulation vs. multi-agent systems
From: Christophe Chevallier (christophe.chevallier@ensmp.fr)
Date: Fri Jul 11 2008 - 15:33:42 CEST
Hello again,
I've had personal answers to my question, that I wish to share on this
mailing list. Actually a paper from Hare and Deadman discusses the topic :
Hare M. & Deadman P. (2004) 'Further towards a taxonomy of agent-based
simulation models in environmental management.' Mathematics and
Computers in Simulation 64 (1)
Christophe.
Christophe Chevallier a écrit :
> Hello,
>
> I have encountered in my readings the terms "Multi-agent system" and
> "Agent-based modelling and simulation (ABMS)" (and even
> "multi-agent-based modelling and simulation"). From a literal
> understanding, I could think the term "multi-agent system" insists on
> the interactions between agents, and "Agent-based modelling and
> simulation" focuses on the individual modelling level as a way to
> describe emergence in complex systems. However, emergence is studied
> in multi-agent systems and interactions are studied in Agent-based
> modelling and simulation. I also couldn't notice in my readings any
> fundamental difference between the theories and concepts they were
> refering to.
>
> Can these terms be used indifferently? When to use one instead of the
> other? Should it be understood that "Agent-based modelling and
> simulation" is a process for studying "multi-agent systems"? Would it
> be equivalent to use "Modelling and simulation of a multi-agent
> system" instead of "Agent-based modelling and simulation".
>
> Thank you for your help,
>
> Christophe.
>