Information Technology Reference
In-Depth Information
2. The Role of Models and Cellular Automata
Obviously, we all have our own “vision” of the future of the city, and that is
obviously a non est disputandum : anyone, each individual, each social class, has right
on its own vision, more or less freely chosen, that then compete and collide with other
visions.
But if (and as long) a vision is accepted and shared, the decision about all the rest
should be defined on the basis of the case in question, it is in that case a true problem .
But in order to solve the problem it is necessary to understand the real situation,
which presupposes the use of instruments and tools for reading, interpretation and
prediction, the use of a “tool-box” with all necessary models.
The role of models in the government of the city is thus central. But which models?
2.1 Characteristics of a Model
We would only like to recall the qualifying elements “good” urban models should
have (maybe not all of them and not immediately):
it should not be a black-box; it is essential that those who use it for planning as
well as those whom the plan is directed to, understand how it works and why;
a model should predict and should take into account actions and reactions of social
actors as well as their interests, conscious or not, disclosed or not, rational or not;
a model should be such as to enable the assessment of as many alternatives as
possible, as well as their comprehension and their differences;
a model should be compatible with other models, even if differing in formulation
and techniques used;
a model should be parsimonious, should not require an excessive number of
variables, an excessive amount of data and an excessive computational power;
a model should be flexible for different situations and contexts, and should permit
to be fed, processed and handled with what is at hand;
a model should be fast to build, at least compared with the time-line of the project
the model is built for;
a model should be re-usable and anyway should never be a hapax legomenon .
Many of the above mentioned characteristics are innate in models based on the
“artificial life” approach, and more specifically based on cellular automata; many, but
not all, and not all to an equally adequate extent.
We are not proposing this technique as just another occasional fashion, or as the
final solution, or as the “right” model, but nonetheless we think it to be a possibly
useful model for coping, in many senses, with the complexity and for obtaining a
good collection of answers.
No more, no less.
Subsequently, we will describe analytically some main elements of a generalised
cellular automata model.
Search WWH ::




Custom Search