Traffic Optimization Platform for Jeju Traffic Safety

AI-driven traffic management system to enhance Jeju's transportation infrastructure using data-driven urban planning

Overview

From April 2024 to February 2025, this project built an AI-driven traffic management system for Jeju Island, supported by the Jeju RIS (Regional Innovation System) Office. The goal was to reduce congestion and improve safety at key intersections by replacing static signal timing with data-driven optimization.

What We Built

The platform collects real-time traffic data from multiple sources and uses machine learning to analyze traffic patterns across Jeju’s road network. Based on these patterns, AI algorithms optimize traffic signal timing at high-congestion intersections, adapting dynamically to changing conditions such as tourist season surges, rush hours, and weather events.

We also developed traffic flow simulation capabilities that allow transportation planners to test “what if” scenarios — for example, predicting the impact of a new road or a major event on island-wide traffic patterns.

Technical Components

The system includes real-time traffic data ingestion, pattern recognition using machine learning, signal timing optimization algorithms, and a simulation engine for scenario testing. All components were designed to integrate with Jeju’s existing traffic infrastructure.

Impact

The platform provides Jeju with a foundation for data-driven transportation planning, with the potential to reduce congestion, improve traffic safety, and support the island’s growth as a major tourist destination. The approach is designed to be scalable and replicable for other regional smart city initiatives.

Collaborators

  • Cheju Halla University: Lead development
  • Jeju RIS Office: Project support and coordination