Water, water everywhere!

A single burst pipe can cost millions, but this new predictive model can save cities significantly.

“Traffic snarled due to a burst watermain near the Granville Aquaduct,” states News 1130.

Due to the estimated age of the infrastructure and pipe material, the City of Vancouver had slated the duct for replacement next year. A year too late for all those affected by today’s rupture.

In Canada, the replacement of pipes in poor and very poor conditions requires a total investment of about 25 billion dollars (Canadian Infrastructure Report Card 2016).

When pipes fail prior to their planned replacement date, the surprise costs infinitely more in unanticipated expenses including overtime and emergency crews, property damage, curtailed traffic, and the unseen costs of lost service delivery, inefficiencies and stress. A single pipe break can cause a domino effect across an entire urban region.

Watermain breaks have a substantial economic, environmental, and social impact on municipalities. So, forecasting pipe breaks and failures through predictive models is essential to assist city water managers in developing proactive maintenance strategies and preventing breaks.

Yet breaks are rarely due to a single factor; rather they result from the accumulation of multiple factors over time.

For instance, water pressure accelerates pipe degradation by aggravating existing problems such as corrosion and micro-cracks. Unfortunately, this factor is rarely considered in current predictive break models, and its impact is poorly understood.

Predictive analytics

That’s where CANN Forecast, a company that applies artificial intelligence (AI) to help municipalities make more cost-efficient and environmentally sound water-management decisions, applies its expertise. Operating from the D3 incubator at Concordia University in Montreal, CANN Forecast produces InteliPipes software — which identifies pipes at risk of breaking.

“The costs to repair pipes can be about $10,000–300,000 in materials and labour, plus indirect costs of pollutions, traffic, etc. A single pipe break can cost millions,” says Dr. Renato Henriques da Silva, a postdoctoral fellow at McGill University and Mitacs intern with CANN Forecast. “Our model will save cities money by better predicting which pipes need repairing now and which others can safely be delayed.”

The software uses the historical data of a city’s water network to learn the complex relationship between the different variables — age, pipe material, and water pressure — that can contribute to watermain pipe breaks. InteliPipes allows municipal decision-makers to optimize replacement activities and better estimate the long-term budgetary needs for the maintenance of the waterworks.

CANN Forecast provides a decision-making tool that helps cities monitor the integrity of pipes to predict failure. The company enables municipalities to move from reactive to proactive decision making, which helps to be more cost-effective and transition to smarter cities across Canada.

CANN Forecast history and team

A start-up founded in 2017 by Naysan Saran and Nicolas Fortin St-Gelais, CANN Forecast traces its origins back to a win at the AquaHacking hackathon with their beach water quality prediction model, InteliSwim. Within the six-months’ time-frame of the hackathon, the team developed the model to address the challenge of delayed test results for urban beaches.

A year later, the Water Network Department bridged the connection from beach-water testing to municipal pipelines and made networking connections for CANN Forecast. The young team next addressed the municipal pipe-failure challenge.

In summer 2019, the CANN Forecast team of seven expanded when they partnered with Mitacs, and appointed Renato, a postdoc supervised by Professors Alexandra Schmidt (Biostatistics, McGill University), and Sophie Duchesne (Centre Eau Terre Environnement, INRS), to accelerate their research.

Born in France and growing up in Brazil, Renato studies break-prediction modelling of pipe networks.

“Sometimes municipalities simply don’t know where to fix,” says Renato. “In light of aging infrastructure, we are developing a model to improve predictions to allocate resources in the right place at the right time.”

With his background both in statistics and in ecology, he’s complemented that study by working as a data scientist, performing data analysis.

Several cities in Quebec including Gatineau, Montreal, and Rimouski, as well as Windsor, London and Thunder Bay, ON and Halifax, NS, have been participating in the project.

Renato explains, that the reasons pipes break include frost, aging, as well as soil corrosion. However, a very important, yet often overlooked problem, is pressure build-up in the system — with too much variation, pressure weakens pipes. Renato is developing a method to model pressure, which is not currently used in modeling predictions.

He adds that this type of modelling can be of tremendous value to municipalities. “For example, the AI model that CANN Forecast developed for Montreal improved their predictions of future pipe breaks more than ten-fold. My research project will improve these number even more.”

“By knowing in advance which pipes are likely to break, cities can pro-actively fix or replace them and thus avoid huge surprise costs.”

With infrastructure aging across Canada, CANN Forecast’s software is replicable for municipalities nationwide.

“We want to create a database across Canada,” says Renato. 

Benefit of Mitacs internship

“My Mitacs internship helped CANN Forecast to have budget to pay me,” says Renato. “The program gave me a unique opportunity to shift my career from academia to industry. Now I am able to apply the skills acquired during my PhD on industry-focused research.”

As for his future, Renato hopes to continue in industry.

“Mitacs has been crucial for our company at the stage we’re in now. We really needed special statistics as well as hydrology. That’s really high-level skills to possess,” says Naysan. “To hire someone without a supervisor would have been a gamble. Having Renato on board, supervised by faculty, reduces the typical research risk.”

Renato is the first Mitacs intern on the CANN Forecast team; and Naysan foresees more in the future.

“Mitacs is a great program. Interestingly, I was in the U.S. speaking with someone who had a start-up. He admired that Canadian startups have such a straightforward access to this kind of academic partnership and funding.”

Mitacs thanks the Government of Canada for their support of the Elevate research fellowship in this story. Across Canada, the Elevate program also receives support from the Government of British Columbia, the Government of Manitoba, the Government of New Brunswick, the Government of Newfoundland and Labrador, the Government of Nova Scotia, the Government of Ontario, the Government of Prince Edward Island, the Government of Quebec, and the Government of Saskatchewan.

Do you have a business challenge that could benefit from a research solution? If so, contact Mitacs today to discuss partnership opportunities: BD@mitacs.ca 

Mitacs Team

Mitacs’s website content is created by people throughout our organization, united in their passion for innovation and eager to share their perspectives with others in the innovation ecosystem. 

More posts by this author Contact Me