Fracta and Toho Gas to perform PoC to predict gas pipeline condition

US-based Fracta has signed an agreement to perform a proof of concept (PoC) using its proprietary algorithm to predict the condition of Japanese utility company Toho Gas’ pipelines.

Toho Gas will use Fracta’s algorithm from next month.

The AI-powered algorithm makes use of more than 1,000 environmental variables to evaluate the condition of the gas pipelines.

Fracta said in a statement: “To increase the efficiency of Toho Gas’s capital expenditure for replacing the old pipes, Fracta will utilise the data collected rigorously by thorough inspections and compilation of data by Toho Gas on their pipeline in line with Fracta’s proprietary environmental database and its cutting-edge AI algorithm used in water mains prediction to refine a new algorithm to predict the deterioration of gas pipelines.”

Once the PoC for the algorithm shows sufficient accuracy after completion this year-end, it will be incorporated into new gas pipeline deterioration prediction software and is expected to be deployed throughout Japan.

Fracta uses AI and machine learning to predict the remaining useful life and replacement time of water main pipes using pipes’ information such as age, material, and year of installation, along with environmental factors that include soil, weather and population.

To be used by more than 40 customers of Fracta in 20 US states, this information will enable them to maximise their capital expenditure on replacing the correct old pipes.

Last month, SUEZ North America signed a partnership agreement to incorporate Fracta’s AI-based condition and risk assessment technology into the delivery of its pipe asset management programmes in the US.