Establishment and Operation of a Data-Based K-Infection Management Support System

Overview

In response to the Itaewon tragedy, it became necessary to preemptively establish a data-driven alert system to prepare for situations involving large crowds. This system would enable local governments to recognize such situations early and support scientific decision-making. By leveraging mobile carrier base station access data and spatial information of the relevant area, it is possible to predict and analyze potential risks arising from crowd congestion in advance, thereby contributing to accident prevention.

Project description

A Crowd Management Support System has been established to collect and analyze data from densely populated areas, monitor risk levels, and provide notification functions. This system collects and analyzes mobile population data* and spatial information such as roads to measure 'density', then displays this as a heatmap on a GIS situation board.
*Data collected from mobile carriers at 5-minute intervals in 50×50m grid (Cell) units

Furthermore, when monitoring detects density levels exceeding the appropriate threshold (Level 3 or higher out of 5 total levels), the risk level is communicated to local governments via situation dissemination messages and to users via mobile messages.

When hosting regional festivals or events such as spring cherry blossom festivals or international fireworks festivals, the Crowd Management Support System can be utilized to deploy safety personnel at event sites, ensuring events proceed safely. Plans include enhancing local government usability by comprehensively considering their needs and operational outcomes while refining the managed areas.

Key technological features

• Providing monitoring by analyzing density based on mobile carrier data
  • Collect and analyze crowd density influencing factors such as mobile population data and public transportation information, displaying them as heatmaps on a GIS situation board
• Providing advance early warning notifications for congestion levels
  • When crowd density risk exceeds an appropriate level (Level 3 or higher out of 5 total levels), provide risk level alerts via situation dissemination messages (to local governments) or mobile messages (to users)

Expected outcomes

The Ministry of the Interior and Safety established a crowd management support system leveraging AI and digital technologies, which began operations at 100 locations nationwide in December 2023. This system monitors potential hazardous situations in dense crowds, enabling swift response and management to prevent disaster safety accidents.

During the Busan Fireworks Festival following the system's implementation, approximately 770,000 people gathered. The crowd management system facilitated the efficient management of this dense crowd by operating integrated on-site situation rooms at both the city and district levels, in coordination with the city's disaster safety situation room, establishing a comprehensive monitoring and response framework.

Before and after implementation (AS-IS vs TO-BE)

Launch year: 2023

Attribution

Case study courtesy of the National Information Society Agency, Republic of Korea, Apolitical's Content Partner. Sourced from Use Cases of Public AI Service, Vol.1,  https://eng.nia.or.kr/site/nia_eng/ex/bbs/View.do?cbIdx=31975&bcIdx=28997&parentSeq=29033