Marco Sordo

Computer Science, Simulation and Economics

Project work on

"A New Interpretation of the Segregation Model "


The applet requires Java 1.4.1 or higher to run. It will not run on Windows 95 or Mac OS 8 or 9. Mac users must have OS X 10.2.6 or higher and use a browser that supports Java 1.4 applets (Safari works, IE does not). On other operating systems, you may obtain the latest Java plugin from Sun's Java site.

created with NetLogo

view/download model file: a_new_interpretation_of_the_segregation_model.nlogo

This "Segregation Model" examines how three populations of different individuals, distributed within a closed system, aggregate into homogenous groups, under the pressure of repeated iterations.

The individuals in the simulation go from one point to another point of the world following a random direction but their behaviour follows an incremental strategy, which means that they tend to maximize their nearness to other units of the same kind by looking for other similar agents. If they find one or more units of the same kind, they stop but if not they keep moving. The model has several controls that decide, first, the number of individuals in the simulation, distinguished in three colors (red, green and blue); a second control is the percentage averaged over the white area, of individuals surrounded by others of the same kind. A third control (the state of aggregation) is the number of possible directions (up to 3) in which the unit is permitted to move, at each iteration.

The first action is to set the number of the units in the simulation, then the user should set the number of the units in the simulation, at this point the user should set the percentage of aggregation that is the goal, starting from a random distribution. Thirdly, the level of rationality of the individuals must be set, defined by its ability to act upon the knowledge of the color of the surrounding units (no knowledge, ie. random movement, or full knowledge, up to 360░).
After the setting procedure, the user should setup the simulation, clicking on the button "setup" and start the simulation, clicking on "go".

After a while, the interface prints the graphics and the monitors, which give a numerical rapresentation of the temporary equilibrium of the model. If the state of aggregation of all the population reaches 100%, the simulation stops. If the model doesn't reach this point, the percentage of aggregation tends to oscillate arund an average value.

If the user creates in addition to the original "Schelling model" two "sliders", consisting in a control of direction and a slider to create a third kind of agents. The simulation is more accurated.

It would be interesting to study the model by adding a third dimension in heigth. It is more difficult to calculate the equilibrium in this case.

This model creates a third group of patches (the blue ones) and also uses a new global variable, called direction to measure the level of rationality of the agents.

A model that I have used to understand how the Schelling Model calculates and searches the equilibrium is the "Artificial Financial Market" model.


Schelling, T. (1978). Micromotives and Macrobehavior. New York: Norton.

See also: Rauch, J. (2002). Seeing Around Corners; The Atlantic Monthly; April 2002;Volume 289, No. 4; 35-48.

Wilensky, U. (1998). NetLogo Segregation model. Center for Connected Learning and Computer-Based Modeling, Northwestern University, Evanston, IL.


P.Terna, R.Boero, M.Morini, M.Sonnessa (2006), Modelli per la complessitÓ. La simulazione ad agenti in economia. il Mulino, Bologna. Look at