WIREDGOV

Blog posted by: Atos, 15 October 2021.

By 2022, more than 50% of enterprise-generated data will be created and processed outside the data center or cloud[1]. By sending only the most important and least time-sensitive information to the cloud, edge computing eases the burden on the cloud and reduces costs[2].

Industrial IoT…but moving to the Edge Leveraging on hardware agnostic Codex Smart Edge smart node software

Waste water processing Applied case study of edge and AI

Wastewater treatment is complex. As plants are measured on the quality of their water after treatment, they cannot afford to take any risks. The process of cleaning water (for example removing sludge and moving large volumes through the plant) combines the use of large amounts of chemicals and energy. Being such a power intensive business means major demands on the grid and potential penalties for over consumption

There is opportunity in every step of the treatment process to reduce energy use, improve efficiencies and reduce cost. In this scenario the plant needed a more efficient way to be able to collect and aggregate huge volumes of data, either by sensors, camera or both to make more informed decisions.

Using advanced analytics and our big data platform, we identified the biggest pain points and therefore, the areas to optimise. Using Machine Learning, we then developed algorithms to be trained centrally in the cloud and executed in real-time on site using Codex Smart Edge. The trained algorithms were packaged in Docker components, downloaded to the edge and run on site using our Codex Smart Edge solution. This uses both real time IoT data collected on the plant as well as third party data. Analytics in Codex Smart Edge meant a co-pilot facility could be offered to the plant operator, as well as real-time decision making in an automated closed control loop for plant optimisation.

Water treatment optimisation

Analytics roll-out all along the value chain

Key Takeaways Position for the Post-Cloud Era

Edge Computing is a massive revolution, enabling compute and real-time analytics close to the data sources, outside of data centres.

Data is becoming more and more complex, making AI and Machine Learning critical technologies to analyse massive volumes complex data sets.

[1] The Edge Completes the Cloud: A Gartner Trend Insight Report (09/18).

[2] Arriving at the edge of cloud computing A TBR Special Report (10/19).