CANN Forecast and the City of Montréal
CANN Forecast (formerly Info-Baignade), winners of the 2016 AquaHacking Challenge, recently signed a 4 month contract with the water department of the City of Montreal for a pilot project to determine how machine learning can help with the prediction of water main breaks.
The goal is to allow the water department to better target preventive maintenance tasks on their drinking water infrastructure by conducting a comparative analysis of the performance of several machine learning algorithms to predict historical water main breaks for each district, based on data provided by the City.
CANN Forecast is hoping to collaborate with a few other cities in Canada and/or the US in 2018-2019 in order to evaluate how their machine learning models perform when trained with their data. The ultimate goal is to provide municipalities that face the same challenge as Montreal with a turnkey solution that includes every step from data analysis to a graphical interface that displays which pipe segments are most vulnerable to failure.
More info on CANN Forecast and other winning solutions in development: https://aquahacking.com/en/joi...