AIOI R&D LAB-OXFORD, in collaboration with Aioi Nissay Dowa Insurance, started the joint development of a bridge inspection support tool. This tool will detect damages and quantify the health of bridges managed by Susono City using inspection photos.

Background

Infrastructure aging is a significant global issue. In Japan, much of the infrastructure was built during the high economic growth period of the 1970s. Approximately 59% of the country's 730,000 bridges will have surpassed 50 years of age in the next decade, a critical threshold for preventive maintenance according to the Ministry of Land, Infrastructure, Transport, and Tourism. Local governments face challenges in bridge maintenance due to a lack of technical staff and budget constraints, necessitating efficient maintenance methods. In response, Aioi Nissay Dowa Insurance established Aioi R&D Lab -Oxford (Lab) in November 2022 to develop advanced AI technologies. Together with Susono City, which is striving to create a city chosen by people and companies using advanced technologies and data, the company aims to enhance the efficiency and sophistication of bridge inspection processes using AI.

Overview of the Joint Development

The bridge inspection support tool consists of two main functionalities developed in collaboration with the Lab:

1. Detection and Quantification of Bridge Damage (Expected Completion: February 2024)

   1. Function: Uploading photos of damaged parts of the bridge taken during inspections to the app will enable the detection and quantification of damages like cracks and rust.

   2. Data & Development Process: Analysis of 297 bridges managed by Susono City and photo data of inspections for AI learning.

2. Automatic Assessment of Bridge Health (Expected Completion: July 2024)

   1. Function: The app will evaluate the bridge's damage state from photos and automatically determine its structural health.

   2. Data & Development Process: Analysis of past bridge inspection data and health assessments from Susono City based on the Ministry of Land, Infrastructure, Transport, and Tourism's guidelines for AI learning.

Application Provision

The app, equipped with these two functionalities, is scheduled for development by July 2024. Local government staff or inspection contractors can upload bridge images to the app for damage detection, quantification, and structural health assessment. The app will also automatically generate inspection reports based on AI-detected and assessed results, reducing the workload of measuring damages and standardizing health assessments, thus contributing to accurate and efficient bridge maintenance.

Future Developments

Aioi Nissay Dowa Insurance plans to roll out the developed bridge inspection support tool nationwide to local governments from January 2025. The company also aims to advance AI using telematics data, such as vehicle vibration data, and apply AI in the maintenance of infrastructures other than bridges. Susono City intends to use this initiative to build an efficient bridge inspection system, accumulate high-precision judgment technology and know-how, and promote safe and forward-looking city development, as well as to realize citizen services suited to the times and needs.