Machine Learning now available as short-term step to Net Zero

Partner Content

In this article we drill into why Monitoring & Targeting often doesn’t deliver on its true savings potential and how Machine Learning can overcome this.

Net Zero has never been a bigger topic. It’s everywhere and rightly so, it’s one of, if not the greatest challenges of our generation.

Amongst the sea of Science Based Targets and long term reduction strategies, it’s clear that there is no single answer to reaching net zero and for most organisations it’s a marathon not a sprint. However many businesses are looking again at what they can do in the short term to make an immediate impact.

Reducing demand is the first step of any energy hierarchy and utilising data in a commercially-viable way is critical to this. Many organisations understand the merits of measuring the enabled savings but not all businesses follow through on this.

Many users of Monitoring & Targeting (M&T) platforms will know about the benefits but also the challenges that exist in extracting value from their system. The market is now flooded with a broad range of M&T systems, with varying degrees of flexibility on data collection, analysis tools and reporting outputs. However all such systems are constrained by the time available for skilled people to review the data and identify areas for improvement. In fact, this problem is now greater than ever as systems become more interconnected and data availability increases.

Production teams want to do the right thing but time is the biggest constraint on effective energy management

Verco has been working with many of the UK’s leading manufacturing businesses and observed this growing issue over the last 15 years. Over the last 5 years in particular, there has been a rapid acceleration in data availability coupled with increasing limitations on available man-power to make use of this. Data availability is no longer the number one constraint as it was a decade ago.

We’ve taken collective feedback from hundreds of manufacturing sites, all trying to do what they can on energy management surrounded by the real-world constraints that exist. We’ve seen countless examples of stretched site teams unable to utilise the data they currently have access to. This, in turn, makes it very difficult to make the business case for further automatic data collection of the more complex systems which is often where the greatest hidden savings exist.

We’ve concluded that this challenge isn’t going to go away and have taken a new approach utilising AI

Verco has been using the very latest Machine Learning tools to create a new generation of system that is befitting of the next decade. We’ve already invested many thousands of developer hours into testing our hypothesis that machine learning can and should play a significant role in solving this problem. We’ve seen some hugely positive results from our trials and Early Adopters Programme, with the trained algorithms identifying issues that would otherwise require the human eye to spot, as well as trends that go beyond what the human eye would see and would otherwise be missed.

Reduct combines flexible data collection methods, with a bespoke application of machine learning algorithms developed by Microsoft and others to provide you with the best available solution for AI-driven energy management. The system works in a similar way to other M&T platforms, but significantly reduces the need for user data review as AI does this for you. This removes a time-consuming step which is often the stall point of standard M&T systems.

Find out more about Reduct

Visit the Reduct website here to find out more about this exciting innovation to transform your data into insights.

Watch the Reduct walkthrough

See a demo of Reduct in action here, the dashboards and reporting outputs. Get a feel for what Reduct can do for you!


About the author

Tim Kay, Director at VercoTim Kay, Director at Verco

Tim is a Chartered Engineer and Chartered Energy Manager and has 15 years’ experience at both operational and strategic levels utilising data to reduce carbon emissions.