Kashima Corporation and Ridge-i have created an AI system that automatically detects problems during concrete filling in CFT pillars—structural columns used in building construction. The system analyzes real-time monitoring data to spot defects like voids and improper flow timing, eliminating the need for constant manual oversight and improving construction safety and quality.
Summaries like this, in your inbox every morning.
Sign up free →What happened
Kashima Corporation and Ridge-i developed an AI system that detects anomalies during concrete filling of CFT (concrete-filled steel tube) pillars. The system uses AI to analyze data from the monitoring system "moni-as" and identifies issues like voids, cracks, and improper filling timing in real time.
Why it matters
In CFT construction, monitoring concrete filling quality is critical for structural safety. Previously, site workers needed constant on-site oversight to spot problems. The new AI system enables automatic, objective detection without relying solely on worker experience, reducing the burden on skilled technicians and improving quality assurance on construction sites.
What to watch
The AI model was trained on over 2,000 concrete filling records from moni-as monitoring data, combined with site knowledge and expertise. The system checks three key points—pouring timing, concrete flow, and diameter timing—and produces a three-level risk assessment (green, yellow, red) that workers can see in real time. Full-scale deployment is planned for summer 2026.
CFT construction requires careful control of concrete filling to ensure structural integrity, but detecting quality issues has relied heavily on experienced site workers. Kashima's collaboration with Ridge-i addresses a real pain point: the shortage of skilled labor and the difficulty of maintaining constant, objective quality oversight on busy construction sites. By training an AI model on historical monitoring data paired with expert site knowledge, the system can now make the same judgment calls automatically and at scale.
The system integrates with moni-as, an existing real-time monitoring platform that already tracks concrete filling conditions. By layering AI-driven anomaly detection on top of this data stream, the companies achieve real-time alerts without requiring workers to interpret raw sensor readings. The three-point assessment (timing, flow, diameter) maps directly to the most critical failure modes in CFT filling, suggesting the model is designed to surface the anomalies that matter most to construction safety. The planned 2026 rollout indicates the system has passed internal validation and is moving toward commercial deployment.
No comments yet. Be the first to share your thoughts!
Log in to join the discussion
Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.
Get Started FreeFree · takes 30 seconds · unsubscribe anytime
1 minute a day. The AI essentials.
200+ sources · Email / LINE / Slack
