Background

LG Electronics operates commuter buses on long-distance routes and identified the need for AI-driven solutions to improve driver safety, prevent accidents, and reduce risky behaviors. Long routes present higher risks of drowsy driving, inattention, and momentary driver errors that can lead to accidents.

To address these risks, LG partnered with the Korea Transportation Safety Authority (TS) to pilot A.I.MATICS’ Vision AI–based driver monitoring solution. The system detects risky driving behaviors in real time, sends alerts, and provides accident response support.

Solution

From September to November, A.I.Matics’ AI safety solution was deployed on 14 LG commuter buses serving routes between Changwon and Busan.

Key features included:

• Real-time detection of risky driving behaviors and instant alerts
 (drowsiness, distraction, smartphone use, smoking, seatbelt neglect, etc.)

• Blind-spot detection to prevent collisions

• Instant accident alerts with automatic location data sent to managers

• Driver behavior monitoring and detailed reports for safety analysis

The Roadscope AI dashcam was equipped with 5 cameras (front, rear, cabin, left, and right) to provide full monitoring of both driver and vehicle status in real time.

Results

The pilot demonstrated significant improvements in safety performance:

• 70% improvement in driver safety scores (48.4 → 82.3)

• Major reduction in risky behaviors:

• Phone use: ↓44.4%

  • Smoking: ↓46.3%

  • Drowsy driving: ↓80.3%

  • Distraction: ↓64.2%

  • Inattentive driving: ↓56.7%

• Zero drowsy driving incidents recorded (vs. multiple cases before)

• 43% reduction in bus idling time, improving fuel efficiency and cost savings

Based on these results, LG Electronics plans to expand adoption of A.I.Matics’ AI safety solution across all commuter buses by 2024. The initiative is part of a broader effort to create safer commuting environments, reduce driver fatigue risks, and achieve measurable economic benefits.

The Korea Transportation Safety Authority also plans to use this project as a model to expand AI-driven safety monitoring solutions to nationwide fleets, in collaboration with the Ministry of Land, Infrastructure and Transport, municipalities, and private operators.