How to solve problems with sparse modeling (Part 6)

How To Solve Problems With Sparse Modeling (Part 6)

Hello everyone, My name is Kenshin Fujiwara and I am the CEO and founder of HACARUS Inc. 

Through this series of blogs, I will discuss a wide range of topics of AI, from the history of AI to practical tips for the successful application of AI projects. I hope my blog posts will help you better understand AI and solve your business issues.

In the last blog, we studied a case study of AI implementation for diagnosis support at Kyoto University. Today, we will take a look at a case of AI implementation at Osaka Gas, a Japanese utility company. 

AI to Locate and Map Buried Underground Pipes

In this case, Osaka Gas, a Japanese utility company, introduced sparse modeling AI to locate and map buried underground pipes. To accomplish this feat, a device called ground-penetrating radar was used. 

Looking at the figure, you can see the cart that is operated by a worker who walks along the road. While in operation, the radar emitted from the device captures the reflected waves bouncing off the buried pipe. 

These waves are then displayed on the screen of the laptop mounted to the device. Since the radar output is difficult to read, it takes a trained eye to decipher the display. 

By looking at the outputs, the field crew is able to tell the location, depth, and type of piping. It is also worth mentioning that gas pipes are not the only thing buried underground. There are many other sensitive utilities buried underground including water and sewage pipes as well as power and communication cables. 

Of course, Osaka Gas has information about the location and size of gas pipes that it had buried previously. However, there are other buried objects that were installed by other companies. Before starting any roadwork or construction, the area must be surveyed so this information is extremely important. 

In the worst-case scenario, an object might be damaged during the excavation, and the social impact would be significant. This outcome could lead to the shutdown of infrastructure in a wide area. This further demonstrates the value and importance of this technology. 

These surveys of buried pipes are conducted on a daily basis over a wide area. Since Osaka Gas provides services across the Kansai region, there is a high demand for an easy-to-use method for locating buried pipes. 

Industry Challenges Prior to AI Implementation

  • Surveying required skilled workers to decipher images.
  • The aging of the workforce has created a shortage of workers
  • The skills of operators are highly individualized

Again, the main objective of introducing AI was to take over this humanized exploration capability. To accomplish this, it is important for AI to display results in real-time in order to introduce them into the field.

AI Achieved Higher Detection Rate than Manual Inspection

Therefore, Osaka Gas aimed to create a system where anyone could easily conduct accurate, real-time surveys on-site simply by running the AI program. This system was also able to run offline. In the end, the AI achieved a detection rate of 89% which was a 10% improvement compared to the conventional manual inspection methods. 

Once again, the main obstacle for this project was the data barrier. In many cases, buried pipes cannot be identified without collecting data by digging directly into the ground. Also, since the results vary depending on the soil composition, a large amount of data is required which is both time-consuming and costly for the company. 

This is where sparse modeling, which enables AI development even with small amounts of data, came into the picture. In the case of ground-penetrating AI, sparse modeling is a perfect fit since it can make accurate judgments while running on limited computational resources. Because surveying is performed outdoors, it is important that the AI system is portable and compact. 

There is also a high demand for this type of service. Aside from Osaka Gas, other infrastructure companies and railroads regularly conduct surveys to detect buried objects when laying new utility lines or rails. Municipalities also routinely conduct cavity surveys underneath roads. 

In the next blog, I will discuss differences in the use of AI between Japan and the rest of the world. For your updated knowledge and insight about AI technology, subscribe to our newsletter or visit HACARUS website

Subscribe to our newsletter

Click here to sign up