We built the city of Colombo in Cities:Skylines

Original Article ↗ Hacker News Discussion ↗

Project Goals and Use Cases

  • Built Colombo in Cities: Skylines as a “toy” digital twin around 2020 land use, zoning, and transport data.
  • Intended as a visualization and communication tool for citizens and students, not a professional planning model.
  • Target applications: illustrating transport plans (e.g., COMTRANS), showing effects of road and policy changes, teaching complex urban systems.

Accuracy, Limits, and Game Mechanics

  • Repeated emphasis that this is not a fully accurate simulation; closer to a teaching / awareness tool.
  • Skeptics question whether a commercial city-builder with simplified mechanics (no rush hour, weak parking model, gamified social systems) can approximate real problems.
  • Authors acknowledge major constraints: vehicle and agent limits, overly efficient roads due to fewer cars than reality, perfect transit schedules, coarse behavior models.

Methodology, Mods, and Technical Work

  • Heavy use of mods to adjust traffic AI, lifecycles, population density, parking, and South Asian driving norms (e.g., reckless drivers, U‑turns, lane behavior).
  • Many automated imports (heightmaps, OSM roads) failed or were mis-scaled; large portions of the network were redrawn manually using image overlays and coordinate tweaking.
  • City runs near engine limits with modded caps for area, citizens, vehicles, and routes.

Real-World Reception and Applications

  • Urban planners and university departments show strong interest, mainly for teaching and exploratory scenario work rather than formal policy decisions.
  • Students and transport/urban design academics want to apply the approach to other Sri Lankan cities and higher-fidelity subareas.

Cost, Tools, and Open Source Alternatives

  • Cities: Skylines ($20 locally) seen as vastly cheaper than professional tools like CUBE/OpenPaths ($6k–8.6k/year).
  • Open-source 3D/traffic tools exist (e.g., 3D street visualizers, AB Street) but don’t yet match CS’s scale + fidelity combo.
  • Some argue a custom open engine would be ideal; others note this requires years of work and major funding.

Public Funding Debate

  • Some question EU-funded support for what looks like a game project.
  • Others argue it’s a small piece of a broader package (mapping, ag sensors, media literacy) and justified as public education, participation, and research infrastructure.