MLIT's Inspection Support Technology Performance Catalog — Japan's National Certification System for AI Bridge-Damage Detection

Facing a legal requirement to visually inspect every road bridge and tunnel every five years, and a projection that around 40% of Japan's bridges will be over 50 years old, Japan's Ministry of Land, Infrastructure, Transport and Tourism (MLIT) launched the Inspection Support Technology Performance Catalog in February 2019. The catalog lists AI image-recognition, non-destructive testing, and measurement/monitoring technologies — including AI systems that automatically detect and classify cracks, spalling, rebar exposure, water leakage and lime efflorescence on bridge concrete surfaces from photographs or drone orthomosaic imagery — alongside each technology's declared accuracy and applicable scope, so local governments and engineers can choose vetted alternatives to costly close-range manual inspection.
The catalog has expanded from 12 listed bridge technologies in February 2019 to 215 by April 2025, split across image-measurement (91), measurement/monitoring (76) and non-destructive testing (48) categories, with new technologies added through periodic public announcements. From fiscal year 2022, use of catalog-listed inspection support technologies became mandatory for specified inspection items on national-expressway bridges and tunnels, and the requirement was extended to pavement inspections from fiscal year 2023 — turning a voluntary reference list into an enforced national compliance mechanism.
Individual technology performance sheets (such as the BMStar AI damage-detection system) publish declared accuracy claims, including assertions that AI-based crack recognition matches or exceeds close-range visual inspection by trained engineers; these are developer-submitted performance declarations reviewed against MLIT's standard test items rather than fully independent third-party accuracy audits, so the strength of evidence per individual listed technology varies.

Read the full analysis: https://www.mlit.go.jp/road/sisaku/inspection-support/