Business Value and Use Case
- Solution Story
It is imperative to assess water quality on an on-going basis for water distribution networks. The level of impairments need to be checked and appropriate actions need to be taken at the right time in case there is a breach. All of this is made possible only if there is a comprehensive view on water quality index as a whole as well as individual chemical concentrations and their respective levels at any given point in time in the water.
Wipro’s Water Quality Management solution attempts to help organizations control and govern the WQI and respective chemical concentrations so it is in compliance with set standards and does not harmfully impact society and/or environment.
The key features include :
- Monitor water bodies, nearby ground characterises, air and any other influencers of water quality through sensors IOT devices
- Gather historical data to determine the pattern of water quality, events before, after and various other factors
- Predict upcoming water quality which can lead to root causes and take preventive measures to reduce or prevent water contamination
This solution is built using the IBM Cloud Pak for Data (CP4D) AI platform. This Data & AI Platform supports:
- to integrate all the data from IoT sensor data, environmental data, historical data in database
- to become faster in making decisions through intelligent ML based algorithmic insights on Water quality and corresponding chemical concentration
- end to end build support that allows easy collaboration, import, tweaks, deployment and embedding of models and insights
- driving customer value through cost saving and ensuring compliance for utility providers
How Did we do it ?
The following components were involved
Data Virtualization : A framework that allows for data consolidation and merging at the source systems by creating virtual copies of the data and not having to move the data loads in-memory. The table joins and data transformation is carried out here for all input raw data to make it analytics ready.
Project space : this consists of data assets (the prepared and transformed data) which is ready to be fed to the ML model and the notebooks (where the codes for threshold setting and predictions) exist.
The first bit of code as seen, calculates the WQI values and upper limit, while also preparing the data to be consumed by the python code for prediction.
The next bit of code works the actual prediction through for the next 10 days. The model used in the ensemble ARIMA model (at site and chemical level), since it picks out the best combination of variables and has the flexibility to make changes.
Dashboard : The insights are then presented to business users through Cognos dashboards.
Water Quality Index – This gives a geo spatial view of WQI across different zones. The color indicates the average levels of WQI indicating the problem areas in red vs the normal ones in blue. The size indicates the average levels of WQI ranging from lower to higher values.
Water Quality Index trend view – This view shows the details of Top 5 sites which are doing well on WQI to start with. The category wise distribution has all sites bucketed into the excellent, good, fair, marginal categories basis their WQI levels. A deeper view of site wise WQI with the upper limit is depicted below which shows how WQI levels are within breach ranges. The right hand side shows a year on year view with WQI across sites for good or bad performance (color coded).
Chemical concentration – As the name suggests this shows a view of individual chemicals and the respective concentrations that they are present in (mg/L). the top best performing chemicals (site wise) and the poor performing ones are plotted on the charts. The trend over time for individual sites and their chemical levels are seen on the right hand side. The patterns are seen of peaks or troughs over time for the sites.
The last part of the dashboard shows the historical levels of WQI for each of the chemicals captured over last 10 days and the ARIMA based predictions for future WQI levels are seen on the right hand side. This will help water utilities take proactive actions to keep the WQI under check by taking the necessary actions.
Prerequisites of Cloud Pak for Data Services
- Data Virtulization
- Watson Machine Learning
- Watson Studio
- Cognos Analytics
- Please refer here - Water Quality Management
Import the Accelerator
- Please watch the video available in the box folder for the use case. For any questions or clarifications, please contact email@example.com
-This has been tested with CPD Version 2.5
About the Developer
Wipro Limited (NYSE: WIT, BSE: 507685, NSE: WIPRO) is a leading global information technology, consulting and business process services company. We harness the power of cognitive computing, hyper-automation, robotics, cloud, analytics and emerging technologies to help our clients adapt to the digital world and make them successful. A company recognized globally for its comprehensive portfolio of services, strong commitment to sustainability and good corporate citizenship, we have over 160,000 dedicated employees serving clients across six continents. Together, we discover ideas and connect the dots to build a better and a bold new future.