A post by Meniscus Systems Ltd
For Meniscus, being involved in the Event Duration Monitoring (EDM) proof-of-concept trial has provided us with an opportunity to test and enhance our MAP Rain service in a collaborative and solutions-focused environment.
MAP Rain is our cloud-based service combining radar and forecast rainfall with operational data. Using machine learning, we use historical data to characterise each asset – i.e. a combined sewer overflow (CSO) – which allows us to identify performance in real-time as rainfall and CSO alert data becomes available. We use these insights to generate a range of alerts that are e-mailed or made available via the MAP Rain API. These insights could also reduce alert volumes in the control room as we can classify when an asset is operating as expected during a rainfall event. During the proof-of-concept, we were fortunate enough to have some varied periods of rain so we could review how MAP Rain worked across a broad range of rain events.
Overall in the period from the 1 June 2020 to the beginning of 6 August 2020, we identified 90 separate alerts for a range of ‘Dry Weather’, ‘Post Storm’ and ‘Inferred Blockage’ events. Over 90% of the CSO spill data received from the Wessex Water telemetry system was classified as ‘normal operation in wet weather’ and represents the times when the CSO is operating as expected. We reviewed each alert and confirmed its validity and the operational impact/significance to Wessex Water.
This review process identified a number of small changes that we implemented straight away, i.e. fixing data timing discrepancies, but a few changes are more significant and have involved quite a bit of time in the development, testing and deployment of solutions. An interactive proof-of-concept like this is an excellent catalyser to this process.
On a more strategic level, we have had the opportunity to review, discuss and challenge the efficacy of these assets, and the insights we generate, with the Wessex Water team. This has been invaluable as it gives us a better understanding of how this data will, or will not, be used to achieve reductions in pollutions. As a consequence, we have implemented changes in our service that have both reduced the number of alerts generated whilst improving the validity of the alerts that we do send out to the Wessex Water team.
The overall concept behind the Wessex Water Marketplace is one that we are wholly supportive of. The challenge itself was clearly defined. The review process was thorough and the innovation to pay suppliers for the proof-of-concept is so important for smaller companies and demonstrates Wessex Water’s intent to engage with the process. Maybe these factors are the reason why a large number of companies applied to this, and other Marketplace challenges.