In 1971, Thomas Schelling invented an agent-based model, which explains how racial segregation works. This model demonstrates that even little preference leads to total segregation in the long run. He used different coins and placed them randomly on a physical board. Then, he moved all the ones who were in an "unpleasent/unhappy" situation until everyone was happy. Have a look at this short explanation:
What does this model show? First of all, groups are not intrinsically racist (but of course, some members can be). However, already little preference can lead to total segregation. Secondly, the initial conditions are important: An already segregated area does not just change to being an integrated one. Thirdly, there is a tipping point, which means that in case a few households of one race move away, it can be enough that others go somewhere else too (consider also "white flight") and this leads to segregation.
This model can also be applied to other features than skin colour. For example, consider a mixed school class at their first day: Should the pupils decide where they can sit (probably you will get clusters) or should there be an alteration of boys/girls, which would improve the interaction between the genders?
I particularly like Schelling's model because it is simple, yet so witty. Of course, in the real world there are many other important factors we have to consider as well, but it can be at least a starting point (and something to show off with?).
- Schelling, Thomas C. "Dynamic Models of Segregation." Journal of Mathematical Sociology 1 (1971): 143-186. Print.