Cellular Automata Summary

Cellular automata represent a grid-based set of simulations that can be used to represent traffic flow, urban growth, colonies, and crime patterns. Cellular automata are a class of Self organizing ecosystems that evolve over a number of time steps according to sets of rules. The rules for a cell changing state typically involves the state of neighboring cells.

The theory and practice of cellular automata is rich in the results produced. A simple grid and a simple set of rules can produce complex and unexpected behaviours. Thus cellular automata can model complex behaviors with simple rule sets. And cellular automata represent the synthesis of global behavior based upon strictly local rules and behaviors.

Cellular automata can be generated on a one-dimensional grid or a two-dimensional grid. Two-dimensional cellular automata can be used to form landscapes, predict urban growth, and/or generate crime patterns. The figure below shows the result of a typical two-dimensional cellular automata representing a landscape for warfare planning.

Contributions to Body of Knowledge

In this work, I utilize a cellular automata based upon police reporting districts and their associated historical crime levels to model and predict crime levels for a medium sized city of approximately 500,000 people. In addition, the effects of policing and public “targeting” of crime levels is explored for their effects on predicted crime. Read the Medium article here.