| method |
purpose |
mathematical foundation |
description |
level |
# levels |
attributes |
time |
state change |
communication between agents |
complexity of agents |
# agents |
applications |
cons |
software |
| Systems dynamics and world |
prediction |
differential equations |
- describes the target system as a single entity or object: a system dynamics model is an indivisible whole
|
macro |
1 |
continuous |
approximating continuous time |
deterministic |
no |
low |
1 |
- hawks, doves, law-abiders
- world models
|
|
|
| Microanalytical (MSM) |
prediction |
Markov chains |
- aggregate level +
lower level[s]
- methodologies
- static
- dynamic
- longitudinal
- use representative available detailed information about the initial state of microunits
|
micro |
2 |
discrete |
discrete, equidistant |
stochastic |
no |
high |
many |
|
- extremely data-based
- very demanding (computing, storage)
|
|
| Queuing |
|
|
- servers +
queues +
customers
(+ agenda)
- discrete event models
- sthocastic
|
|
1 |
|
|
|
no |
low |
many |
- bank
- computer resources (cpu, printer...)
- airport
- public administration
|
|
|
| Multilevel |
explanation |
- sthocastic processes
- synergetics
|
|
multi |
2+ |
continuous, discrete |
continuous, discrete |
deterministic, sthocastic |
maybe |
low |
many |
- opinion formation
- gender segregation
- law-abider (rev.)
|
|
|
| Cellular automata |
|
|
- cells in a regular grid
- few states for each cell
- state changes at each time step
- state = F(previous state, neighbourhood cells)
- local interactions
|
|
2 |
|
|
|
yes |
low |
many |
- the game of life
- gossip
- majority
- Axelrod's tribute
- migration
|
|
- Game of Life
- (LISP implementations)
|
| Multi-agents (distributed AI) |
explanation |
- (rules) |
- agents properties
- autonomy
- social ability
- reactivity
- proactivity
- problems
- fragility
- complexity
- lack of common sense
- solutions
- production systems
- object orientation
- language parsing and generation
- machine learning
|
multi |
2+ |
discrete |
discrete, event driven |
rules |
yes |
high |
few |
- Sugarscape
- MANTA (ants)
- Evolution of Organised Society (EOS)
|
|
|
| Learning and evolutionary |
|
|
- neural networks
- genetic algorithms
- evolution strategies, evolutionary programming
|
|
2+ |
|
|
|
maybe |
high |
many |
- neural networks
- learning a lexicon
- learning to be altruistic
- genetic algorithms
|
|
|
