Aomori Prefecture, Kajima Develop AI System to Support Bridge Inspections
Damage Detection, Condition Assessment, and Health Diagnosis
November 18, 2025 – Technology / Products
(Image: Example of regular bridge inspection using the new system – from press materials)
Aomori Prefecture and Kajima Corporation have developed an AI-assisted system for regular bridge inspections. The system uses AI to analyze images of damaged areas taken by inspection engineers, supporting the detection and classification of damage types and severities, as well as health-condition assessments. This enables accurate and consistent diagnostics without variations caused by differences in individual engineers’ skills or experience. By promoting adoption of the system, the shortage of inspection engineers can be alleviated while improving inspection accuracy.
The new system, “BMStar_AI”, partially incorporates functions from BMStar, the bridge asset-management support system jointly developed in 2006. Utilizing AI in combination with the extensive inspection data accumulated in BMStar, the new web-based system provides comprehensive support for bridge inspections.
Bridge inspection data collected over many years by skilled engineers using BMStar are used as training data for the AI. During regular bridge inspections, engineers can take photos of damage using smartphones or tablets, and the system immediately displays diagnostic results. For concrete bridges, the AI detects cracks, spalling, exposed rebar, and water leakage or efflorescence from the captured images, classifies the damage, and instantly outputs the health assessment.
For steel bridges, the AI directly performs health diagnosis from the damage images and displays the results. Multiple images taken with cameras or other devices can also be uploaded and evaluated in batches.
The new system is currently being used in regular bridge inspections managed by Aomori Prefecture. Damage detection and classification for concrete bridge spalling, exposed rebar, and water leakage/efflorescence are carried out based on the Ministry of Land, Infrastructure, Transport and Tourism (MLIT) guidelines for regular bridge inspections, while health diagnosis follows Aomori Prefecture’s evaluation criteria. Because the damage detection and classification standards conform to MLIT’s inspection guidelines, the system can be applied to regular inspections of concrete bridges nationwide.
The system is sold by Ritec Engineering Co., Ltd. (Minato Ward, Tokyo; President: Atsushi Tanaka), a Kajima Group company.
