700,000 Bridges Need Deterioration Inspection after 50 Years since Japan's Period of High Economic Growth
Bridges are familiar and vital in everyday life as an important part of society's infrastructure. They are built over land or water to serve as roads for transportation. In Japan, many social infrastructures were built before and after its period of high economic growth. According to the Ministry of Land, Infrastructure, Transport and Tourism (MLIT), the number of bridges that have been in use for more than 50 years is expected at least 67% of the total number of bridges (whose construction years are known) by 2033 (*).
In the United States, where road infrastructure was developed earlier than in Japan, deterioration of the road infrastructure became a problem in the 1980s, which was then described as "America in Ruins." The deterioration problem of road infrastructure started to become evident in Japan as well, and periodic inspections were legislated in 2014, which requires periodic checks on 700,000 bridges across Japan once every five years. As a result, the need to carry out inspections more efficiently is greater than ever.
To meet these needs, Fujitsu has considered applying ICT to inspection tasks for bridges to raise the level of maintenance.
- *: The Ministry of Land, Infrastructure, Transport and Tourism (MLIT) "White Paper on Land, Infrastructure, Transport and Tourism in Japan, 2016"
Detecting Abnormalities in Bridges with Acceleration Sensors and AI
Fujitsu invented a technology for verifying the safety of bridge decks with AI. Bridge decks that directly transfer the weight of vehicles travelling over the bridge are subject to fatigue deterioration of bridges and require much maintenance and inspection. Bridge decks are generally constructed as reinforced concrete slabs that are directly subjected to daily traffic loads. Bridge decks are gradually distorted due to vehicle loading and the repetitive distortion causes degradation of structural performance of the concrete and reinforcing steel. In addition, an increased traffic volume and overloaded vehicles can accelerate degradation. As a case study, technology to confirm bridge safety by embedding sensors within bridge decks has been employed. However, since it is necessary to embed sensors within bridge decks, this technology is not available for bridges that have already been built.
Against this backdrop, Fujitsu considered that FUJITSU Human Centric AI Zinrai would help capture more detailed conditions by analyzing data from acceleration sensors.
The Research Association for Infrastructure Monitoring System (RAIMS) (**) carried out verification tests (***) to simulate the situation of many vehicles running over bridge decks and aggregate vibration data from the construction of a new bridge to its compete collapse. Fujitsu embedded strain sensors within the bridge deck for validation to detect the degradation inside the concrete and reinforcing steel. Also, it attached an acceleration sensor on the bridge deck to keep records.
Then Fujitsu analyzed the acceleration data obtained using a data analysis method called TDA (Topological Data Analysis) on Zinrai, and found that when the bridge had developed internal damage such as cracks, significant differences were observed in the geometrical features collected from the vibration data. It was confirmed that the differences agreed with abnormalities detected by the strain sensors embedded in the bridge deck.
This makes it possible to check the soundness of existing bridges by attaching sensors to the bridge decks.
- **: An organization consisting of 14 companies and legal entities, with the goal of promoting the rapid commercialization of monitoring technology. Fujitsu also carries out research and development as a member of the association.
- ***: This verification test was carried out by RAIMS, as part of a commissioned research project on the research and development of technology promoting the use of monitoring technology for societal infrastructure, commissioned by the Ministry of Land, Infrastructure, Transport and Tourism as part of the Cabinet Office's Cross-ministerial Strategic Innovation Promotion Program (SIP), promoting technology to manage, update, and maintain infrastructure.
Making Social Infrastructure Safer and More Worry Free with IoT, AI and Robots
Current inspection tasks are performed mainly by the eyes and ears of engineers. Meanwhile, the Cabinet Office has recently initiated the cross-ministerial Strategic Innovation Promotion Program (SIP), promoting technology to manage, update, and maintain infrastructure with the aim of advancing innovation in social infrastructure management and deploying new technologies.
In addition, the Fourth Priority Plan for Social Infrastructure Development (****) by the Ministry of Land, Infrastructure, Transport and Tourism (MLIT) looks to leverage maintenance technology and develop and advance new maintenance technologies. As part of the Plan, industry, academia and the public and private sectors are working together to innovate infrastructure maintenance for bridges, which is the theme of this verification project, through developing the on-site verification of robots for next generation social infrastructure. In order to ensure low cost social infrastructure that is safe and worry-free, there are high expectations for using ICT going forward.
Incidentally, some people may think that AI's decisions lack bases. Fujitsu also has AI technology that can explain the bases for making decisions derived from deep learning based on existing knowledge (*****). Going forward, Fujitsu aims to develop digital inspection with more advanced AI technology. Fujitsu will continue striving to contribute to safe and worry-free social infrastructure by leveraging AI and IoT.
- ****: "The Fourth Priority Plan for Social Infrastructure Development 2015-2020" by the Ministry of Land, Infrastructure, Transport and Tourism (MLIT) (http://www.mlit.go.jp/common/001104257.pdf)
- *****: Do You Trust AI? Technology to Explain the Basis of AI-produced Inferences (http://journal.jp.fujitsu.com/en/2018/01/23/02/)