Predictive modeling and remote sensing are Waterlix core strengths. WaterliX-Ray, WaterliX-Smart, and WaterliX-LakeWatch are three solutions by Waterlix.
WaterliX-Ray is the X-Ray for water assets,
This machine learning solution visualizes vulnerable locations on a water main like an X-ray demonstrates internal tissues.
WaterliX-Conserve is a predictive model for water leakage inside premises.
WaterliX-LakeWatch is a service for monitoring lakes and areas affected by pollutants using remote sensing and Machine Learning.
Using more than hundred open data sources, proprietary data, GIS layers, and satellite images, this machine learning solution is capable of pin-pointing small segments of a water main that are vulnerable to leak or break. The solution doesn't need any upfront cost i.e. any sensor to be installed, although can import sensor data if they are available.
WaterliX-Ray could successfully identify 85% of water main breaks in 3% of the water network in Kitchener, Ontario (a city with old infrastructure and 250,000 population). The solution is scalable for cities with any size.
Identifying segments within a water main as vulnerable areas reduces the size of the old infrastructure for immediate attention. Cities can change smaller part of their network, reduce costs overtime as well as risks and increase the level of service. This enables cities to have more budget, be more agile and shift it for other important causes to enhance services or accelerate network maintenance/development with no change in their existing budget.
One of the competitive advantages of this model apart from its accuracy is its low demand for proprietary data, any city with base information about their water network can implement this model, since majority of the data are coming from open data sources which are available for any place around the world.
This solution can be used to identify small segments for preventive maintenance as well and is not developed for replacing water mains only.
This solution can identify leakage in a premise based on consumption trends and can accurately identify over 98% of homes with continuous leakage and over 80% accuracy for premises with one day of leakage or more.
This solution enables cities to look at their historical data and map the changes in leakage in different parts of the city to better study the effectiveness of their actions or campaigns for water conservation on each region respectively.
The City will be clustered to separate regions and then the percent of change or leakage levels in each region would be visualized. The map on the right is a report for the city of Guelph using five years of their historical data.
This is a work in progress to identify changes of algae bloom in the great lakes. The goal is predicting the algae bloom and the level of algae presence in each region.
We will report increase or decrease of algae on each side of the lake overtime and will provide bi-weekly or weekly updates.
One other goal is to forecast the movement of algae masses as well as occurance of red tide. Currently there are satellite images to be used for this purpose.
The next step is automating the process to have this report for any lake near a big city anywhere around the world which may need support or request from around the world. Contact us if you are interested: firstname.lastname@example.org