cover
DESCRIPTION
cover. Workshop on Industry and Labor Dynamics. The Agent-based Computational Economic approach. Introduction by Pietro Terna Dipartimento di Scienze economiche e finanziarie G.Prato, Università di Torino [email protected] web.econ.unito.it/terna. _Simulation. - PowerPoint PPT PresentationTRANSCRIPT
October 3, 2003 Wild@Ace, Torino 1
cove
r
Introduction by Pietro Terna
Dipartimento di Scienze economiche e finanziarie G.Prato, Università di [email protected] web.econ.unito.it/terna
Workshop on Industry and Labor Dynamics. The Agent-based Computational Economic approach
October 3, 2003 Wild@Ace, Torino 2
_Sim
ulat
ion
_______________________________________
From the simulation as a (assisted, opaque) mental experiment …
_______________________________________
October 3, 2003 Wild@Ace, Torino 3
mol
ecul
ar in
tera
ctio
n an
d pr
otei
n
• Folding the smallest protein one could imagine: a single alpha helix.
• A complex structure generated from elemetary rules about inter-molecular interaction
helix30.mpeg From www.stanford.edu/group/pandegroup/folding/
folding of an alpha helix using distributed dynamics
October 3, 2003 Wild@Ace, Torino 4
_Age
nt-b
ased
cho
ice
_______________________________________
… to ACE, Agent-based Computational Economics and, more generally,
ABS, Agent-based Simulation
_______________________________________
October 3, 2003 Wild@Ace, Torino 5
inte
ract
ion
amon
g in
divi
dual
s
From www.icosystem.com/
Collegamento a game_launch.htm.lnk
pt
A
B
pt looks at A as its defender at to B as an offender
or
pt has to defend A from B
action
October 3, 2003 Wild@Ace, Torino 6
sim
ulat
ion
and
agen
t mod
els
Social simulation as a computer based way
to execute complex mental experiments,
but also as a via to represent the complexity of real world
simulation = agent-based models
October 3, 2003 Wild@Ace, Torino 7
_How
to u
se a
gent
s
_______________________________________
How to use agents in simulation models:
a radical view
_______________________________________
October 3, 2003 Wild@Ace, Torino 8
a ra
dica
l vie
w
The radical characterization of an ABM must be found into the possibility of real –
direct or indirect – interaction amid the agents, instead of modeling it in a
simplified way with aggregate simultaneous equations
October 3, 2003 Wild@Ace, Torino 9
_A g
ener
al s
truct
ure
_______________________________________
A general structure for agent-based simulation models
_______________________________________
October 3, 2003 Wild@Ace, Torino 10
ERA
http://web.econ.unito.it/terna/ct-era/ct-era.html
NNCSGA
Avatar
October 3, 2003 Wild@Ace, Torino 11
_An
appl
icat
ion:
ent
erpr
ise
sim
ulat
ion
_______________________________________
An application: enterprise simulation _______________________________________
October 3, 2003 Wild@Ace, Torino 12
1-22-138-28-27-7...
market Enterprise front end
units
our jES (a swarm of units)
FE
28
8
27
A system of enterprises and micro productive units
(a swarm)
FE
recipes
ente
rpris
e si
mul
atio
n (1
)Recipes and production units
7
7
October 3, 2003 Wild@Ace, Torino 13
ente
rpris
e si
mul
atio
n (2
)
8-28-27-7
units
FE
28
8
27
FE
7? ?
a
b
c
The orders are placed in the unit waiting lists and executed according to the FIFO criterion
x
we have the phases a, b, c, then
in x we have a choice problem
Recipes on move
7
Avatar, genetic algorithms, classifier systems to experiment and to simulate the effects of decision making:
• experiments in a usual context
• soft computing applications
October 3, 2003 Wild@Ace, Torino 14
the
rele
vanc
e of
our
wor
k
… relevance needed …