Schelling Segregation Model

×

About the Schelling Model

The Schelling segregation model, developed by economist Thomas Schelling in 1971, simulates how individual preferences about neighbours can lead to large-scale residential segregation, even when those preferences are relatively mild.

Each cell represents a location occupied by an agent from one of several groups (e.g., different racial, ethnic, or political communities) or left empty. Agents evaluate satisfaction based on their neighbourhood composition.

How It Works

Each step, every agent checks what fraction of neighbours (within their neighbourhood radius) belong to the same group. If below their tolerance threshold, the agent becomes "unhappy" and attempts to relocate.

Movement Modes

Global movement (default): Unhappy agents can relocate to any empty cell. Faster equilibration but less realistic.

Local movement: Agents can only move to an immediately adjacent empty cell. More realistic but requires higher vacancy to avoid gridlock.

Neighbourhood Radius

Controls how far agents look when assessing their surroundings. Radius 1 = immediate 8 neighbours (Moore neighbourhood). Larger radii mean agents evaluate wider areas, producing different segregation dynamics.

Key Insight

Even when agents are tolerant—satisfied with only 30% same-type neighbours—collective dynamics still produce highly segregated patterns. Macro-level segregation emerges from micro-level preferences that aren't strongly exclusionary.

What to Explore

• Tolerance: Compare 0.3 vs 0.5 vs 0.7
• Groups: Multi-group dynamics differ from binary
• Neighbourhood radius: Larger radii can slow or change patterns
• Local movement: More realistic but slower
• Vacancy: Local mode needs ~25–35%

Reference

Schelling, T.C. (1971). Dynamic models of segregation. Journal of Mathematical Sociology, 1(2), 143–186.