Comment: Curbing industrial energy with AI

Saar Yoskovitz, CEO at Augury, explains the ways in which AI and predictive maintenance can help manufacturers minimise the pain of spiralling energy costs.

With soaring energy prices, thousands of businesses are at risk and on the brink of an impossible choice: close down or make workers redundant. Thankfully, the UK government has announced support to businesses to address the rising energy prices by capping wholesale energy prices for six months from October 1. This support will positively impact all firms, from charities and hospitals to the big high energy consuming manufacturers. Although it is currently unknown how much this support to businesses will cost the government, it comes on top of the £150bn plan supporting UK households. While this is a welcome decision, any further delay could have proven deadly for many manufacturers, and no business looking to be sustainable long term should be reliant on continued government support.

A deadly delay averted

The latest research from think tank Red Flag Alert found 355,000 high-energy -using companies were at risk of imminent insolvency or likely to lay staff off, with a third predicted to be insolvent by the end of 2022. Although government support is likely to reduce that number, it is unlikely it will save all 118,000 companies at risk who will disappear according to this estimate. Moreover, most of the companies at risk are in steel, glass, concrete and paper production. They are the heavy manufacturers necessary to support either the light manufacturing sector or vital to addressing other crises, such as housing. And while the government’s support will help them to a certain extent, this support is limited, with the scheme originally only slated to run for six months. 

Taking matters into their own hands

There are a few steps manufacturers can take on their own to future-proof their operations. And now the government has given them support, they have six months to get back on their feet and be ready to not only operate without it, but also be prepared for future challenges, including further potential energy price rises in April 2023. 

One solution for them lies in technology: in sounds, sensors and AI to be more specific. Integrated into the manufacturing process, the data generated from the sounds machines make, captured by sensors and analysed by AI algorithms, can help manufacturers in at least two ways in the next six months. 

Adobe Stock -

First, having foresight into when certain equipment is about to fail provides support to manufacturers and engineers through predictive maintenance. The technology can provide maintenance-related actionable insights and prevent breakdown, as running maintenance before a machine breaks keeps energy consumption steady.

How so? A breaking machine will consume more energy as it will strain to keep up the same output at the same speed as a well-functioning machine, straining it even further and speeding up the deterioration and breakdown. And the startup and calibration process required to get the line back up and running consumes energy but with no productive output, meaning it is essentially wasted energy.

Second, smartly integrated into the manufacturing process as a whole, AI can do much more than support predictive maintenance. It can provide an overview of the whole manufacturing process, highlighting the parts that are running efficiently, but also the ones that aren’t. AI-driven actionable insights in this instance can highlight where in the manufacturing process too much energy is being consumed and could be cut back without any impact to the production line. For example, if you are a bread manufacturer, is it possible to produce the same quality bread by lowering the ovens by one degree? The AI would not only be able to answer this but also provide less obvious tweaks and insights.

Putting a number on savings

It goes without saying that the government’s support can go a long way in helping all businesses and manufacturers covering their energy bills in the next six months. However, well integrated and scaled across the whole production line, AI-powered technology can reduce energy consumption by up to 20 per cent. Turning to real numbers, this could mean a 500 sq ft plant consuming average energy (95.1 kWh in electricity or 536,000 BTU of gas) could save between £1179.3 and £2006.61 a year with the price cap taken into account if counting with electricity and gas prices of Q1 2022 from the government.

A £1000 saving might not sound like much in the context of all the other costs associated with manufacturing, but this is only for one plant. Large manufacturers likely own plants of more than one 500sqft plant, so the number should be multiplied accordingly. The government cap will not last forever and manufacturers should be ready to go back to market prices by April of 2023. And that £1000 per plant can by then be better invested.

Saar Yoskovitz is CEO at Augury