WATER TREATMENT FORECAST

OVERVIEW

Global demand for water has been increasing as urban economic activity expands. A significant amount of effort is required to obtain clean water and the reduction of this effort requirement is a major concern for water utilities. The aim of this project was to predict if water is safe for human consumption based on the chemical composition (such as PH, Hardness, Total solids dissolved etc) for water produced or coming from different sources. This helped the client to make better decisions on whether or not to process the water for consumption.

KEY BENEFITS (MINSKY™)

• User-Friendly, cloud based AI platform
• No coding skills are required for results or predictions.
• Provides you a list of % dependency features that can be used to optimize your business
• Ability to fine tune or optimize the models by trying different algorithms / prediction attributes
• Easy integration with other third party solutions such as TABLEAU for data visualization

RESULTS

• Easy prediction to determine if the water is potable.
• Major cost reductions from good predictions before investment of time/funds trying to purify water with dangerous chemicals.
• Improved efficiency in delivery of good quality water.

EXECUTIVE SUMMARY

Access to safe drinking-water is essential to health, a basic human right and a component of effective policy for health protection. In some regions, it has been shown that investments in water supply and sanitation can yield a net economic benefit, since the reductions in adverse health effects and health care costs outweigh the costs of undertaking the interventions. The objective of this project was to predict if water is safe for human consumption based on chemical composition results so that it will help to achieve regulatory compliance. After a detailed evaluation of their operations, AI Labs (www.ailabs.inc) used its proprietary Minsky™ AI Engine to optimize the models by using a combination of AI algorithms and prediction attributes. In this case, Minsky™ used known water samples (good and bad) and their chemical compositions for model creation to predict water potability from new samples. This solution was optimized and implemented in less than a month.

Typical Water Quality Challenges:

• Lack of accessible potable water requires efficient water purification techniques.
• Lack of low cost purification methods due to harmful impurities such as lead, nitrogen, phosphorous, etc.
• Conventional technology does not allow us to predict in advance if a water sample is approved for drinking purposes.
• Poor monitoring and record keeping of impurities in water samples.

SOLUTION

After comprehensively analyzing the data, we used Minsky™ to accurately model historical data to predict the water potability. This process includes historical data of known water samples (both good and bad) which contains chemical composition data parameters such as PH, Hardness, and Total Solids dissolved which helped us to make more detail analysis on whether or not to process the water for consumption. Once the AI Models were generated by Minsky™ for the selected Algorithms, past known water samples and new water samples data was used by Minsky™ to predict the water potability. Prediction data from Minsky™ was also integrated with 3rd Party data visualization application like Tableau.