IBM’s IT knowledge being used to conserve water

Algorithms developed at Haifa research center can detect burst pipes and even predict breaks.

Water 311 (photo credit: Associated Press)
Water 311
(photo credit: Associated Press)
IBM has recently embarked on a number of pilot projects both here and abroad using smart analytics to conserve water in municipal systems, The Jerusalem Post has been told.
Scientists at IBM Research- Haifa have developed complex algorithms that can help minimize water loss from burst pipes, streamline maintenance and reduce general water loss and even predict breaks, Pnina Vortman, senior manager, research scientist, and Smarter Planet Solutions leader told the Post on Monday.
“Water systems nowadays are getting more sophisticated, containing sensors, smart meters and smart valves,” she said.
With such a system, there’s a constant data flow which can be analyzed to create usage profiles, and regular flow and pressure rates. While many municipalities do not have smart systems, the future is definitely trending in that direction.
IBM isn’t building sensors or smart meters, which transmit usage data wirelessly to a central collection point. What the company is doing is taking the data generated and analyzing it using mathematical models and creating programs to give operators a better sense of what’s going on beneath the streets.
An intelligent system which incorporates smart analytics has several clear advantages, according to Vortman.
“We can detect a burst pipe automatically as it happens and then locate the closest shut-off valve to minimize water loss,” she said, by way of example.
The way it works now, a burst pipe might not be discovered for quite a while and usually would require a municipal employee to fix the problem on site in response to a call-in from a customer. Since IBM’s smart analytics are constantly analyzing the flow and pressure in the pipes and building a profile of every part, any anomaly would be red-flagged immediately.
Similarly, IBM’s programs could assist in water quality control by detecting problems much more quickly and quarantining them.
In general, pipes leak quite a bit. Several years ago, when water was not nearly as scarce worldwide as it is today, people were less concerned if significant percentages of water seeped out into the ground in the distribution process. Now, though, every drop counts.
“Information technology [IT] and algorithms can now detect regular leakage. In the past, employees ran acoustic devices along the pipes to detect leaks. By monitoring use and flows, we can figure out where a leak most likely is,” Vortman said.
Anomalies can also be detected by analyzing the data, and the discrepancies needn’t be large ones. Anything that differs from the consumer profile produced by the analytics would be brought to the operator’s attention.
IBM’s new algorithms could also extend the life of the pipes and reduce leaks to begin with.
“Water networks run on high pressure all the time for firefighters. We can enable dynamic pressure control where high pressure can be provided instantly, letting the system run generally at low pressure. By doing that, the life of the pipes is extended and it also reduces leaks,” according to Vortman.
Predicting pipe breakage would also reduce water loss. By monitoring the state of the pipes, maintenance suggestions could be relayed to build a schedule which wouldn’t inconvenience anyone, she pointed out.
When a pipe bursts, the street needs to be torn up, which affects residents and traffic. If a burst pipe could be replaced just before it wore out, and that maintenance could be done at a time designated by the municipality – like the middle of the night – a lot of hassle could be avoided.
Vortman acknowledged that IBM was not the only one operating in the smart analytics field, but maintained that the company’s long IT history and combination of technologies gave it an edge over its competitors.
Right now, IBM is still proving the value of its analytics through its pilot projects here, in the US, and China and expects to convince potential clients of the value of its programs, Vortman said.
Water is not the only area of focus for IBM’s Smarter Planet Solutions initiative, which was launched about two years ago.
“The assumption is that the world is moving towards becoming more and more instrumented – most things have a sensor of some sort. That means there’s a lot of data to analyze. Moreover, fusing data from several different sources can lead to improved processes,” Vortman explained.
That’s the logic behind the initiative and IBM has decided to focus on six specific spheres where it believes its analytics could be quite beneficial: energy, water, transportation, healthcare, finance and, through its Smart City initiative, safety as well.