Singapore's Ministry of Manpower Uses Data Analytics to Target Workplace-Safety Inspections
Singapore
Singapore's Ministry of Manpower used data analytics to flag high-risk construction firms for inspection: 600 firms flagged in 2019 had …
South Korea · Jinju · See the South Korea profile
Top 50% 60/100 · Ask Evidence Copilot about this practice
Korea Land & Housing Corporation's AI-and-IoT monitoring system flags missing hard hats, falls and danger-zone breaches on 311 construction sites; LH reports site accidents fell from 263 in 2024 to 160 in 2025, about 40% lower.
Korea Land & Housing Corporation (LH), the country's largest public housing and infrastructure developer, has rolled out an AI-based real-time monitoring system named "Neulbom A-Eye" across construction sites and public housing it manages. The system analyses CCTV footage and IoT sensor data (tilt meters, crack sensors) to detect four hazard categories in real time - workers without safety helmets, falls or collapses, falls from height, and unauthorised entry into danger zones - and immediately alerts site managers through IoT-linked warnings.
As of March 2026, LH had deployed the AI analysis solution across 311 construction sites nationwide, with plans to extend it to all LH-managed site CCTV by the end of 2026. LH reported that safety accidents at its construction sites fell from 263 in 2024 to 160 in 2025 - a roughly 40% year-on-year drop - figures the company attributes in part to the new monitoring system, though it did not publish an independent causal evaluation isolating the AI system's contribution from other safety measures introduced over the same period. By August 2026, coverage had reportedly extended toward roughly 350 sites, with the corporation concentrating daily attention on its 20 highest-risk locations rather than spreading resources evenly.
The approach builds on an earlier, separate 2021-22 pilot by the Seoul Metropolitan Government, which tested CCTV+AI hazard detection at around 100 small and medium private construction sites and cut emergency-alert response times from 5-10 minutes to about 30 seconds - an early signal that the broader approach, though run by a different public body, had measurable operational benefits before LH's national rollout.
Read the full analysis: https://biz.heraldcorp.com/article/10701957
Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.
Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.
Where this practice's information was retrieved from, and when.
Singapore
Singapore's Ministry of Manpower used data analytics to flag high-risk construction firms for inspection: 600 firms flagged in 2019 had …
Belgium
Since 2014, Belgium's federal social security service has used the MiningWatch data-mining tool to rank construction, cleaning and hospitality employers …
Chile
ChileCompra cross-checks officials' asset declarations against suppliers' beneficial owners, flagging conflicts of interest before contracts are awarded. In 2025, flagged …
South Korea
Ahead of April 2024 elections, South Korea banned AI-generated deepfakes in campaigning within 90 days of polling. The National Election …
Open full copilot Grounded in cited practices — always check the sources.