Our CEO and Co-Founder, Tom Clayton, recently spoke to FMCG CEO about how AI and ML can boost productivity in manufacturing.

If you missed the original article, catch up below.

The UK’s food and drink manufacturing sector is a cornerstone of our nation’s economy, contributing billions annually and supporting thousands of jobs across the supply chain.

Yet, for many FMCG leaders, the challenge of improving productivity remains front and centre. It’s a long-standing issue that’s holding the sector back from reaching its full potential in an increasingly competitive marketplace.

The good news, however, is that the solution lies closer to home than many leaders might think – the untapped data hidden within the plant and machinery of virtually every factory in the world.

It doesn’t involve breaking the bank with costly factory rebuilds or machine replacements either.

By harnessing Artificial Intelligence (AI) and machine learning (ML), FMCG businesses can liberate the data locked inside existing equipment – such as machine PLCs, drives, and IoT devices – to understand how the overall equipment efficiency (OEE) can be improved.

The result is transformative potential across the entire value chain – increased productivity, optimised energy use, reduced waste, minimised downtime, streamlined processes, and boosted throughput, all while upholding the highest standards of quality and safety.

With the right tools and technologies, this ‘hidden’ data can be unlocked and contextualised – making the previously unknowable, knowable.

This data can then be used to generate actionable insights that transform operations and drive meaningful change.

An Untapped Growth Opportunity

Food and drink is the UK’s largest manufacturing sector, but productivity challenges continue to burden the industry.

But an untapped opportunity is there for the taking, in the shape of £14 billion, according to the latest report from the Food and Drink Federation (FDF) in partnership with strategic delivery consultancy Newton – Future Factory: Supercharging digital innovation in food and drink manufacturing.

However, the report highlights that despite increased investment in the industry, growth rates are lagging behind other sectors.

So where exactly is the pathway to productivity falling down?

While 75% of food and drink business leaders claim to see the benefits of digital technologies, they are reluctant to adopt them because of four major perceived blockers.

First, is what the report calls “a lack of investment in the right places.” Second, the industry fears investing in digital technologies will not yield immediate returns. Third, there’s a worry that the lack of qualified staff to drive the digital transformation poses a serious risk to implementation and delivery. And finally, there’s a belief that legacy equipment will need replacing with expensive new kit and technology to exploit the benefits of digitalisation.

It’s clear that the sector’s leaders think the mountain is too high to climb, but what if these so-called obstacles are more a figment of the sector’s imagination than insurmountable real-world obstacles? Mental blockers rather than physical impediments?

Furthermore, what if waiting anxiously turns out to be the biggest risk of all?

Challenging the AI Misconceptions

The barriers to achieving increased productivity are not as big as the report imagines.

There’s no need for manufacturers to invest in new factories or update legacy machinery, they simply need to tap into the rich seam of data buried inside their existing assets.

AI can extract millions of data points per day – no matter what the make or age of the machine.

Additionally, another common misconception is that AI is causing or will lead to job losses. In reality, it’s increasing opportunity – creating data science roles within engineering and operational teams.

That’s because AI only works effectively when coupled with manufacturing and domain expertise. Engineering teams are needed to tag, code, and teach the algorithm, so it can become self-learning.

AI doesn’t exist to replace jobs, rather it’s revolutionising the way man and machine interact.

A mindset shift is required and so is the acceptance that predictive maintenance and continuous-improvements methods are evolving.

The Positive Impact of AI in the FMCG Sector

Recently, a major player in the sector – and one of our customers – implemented an OEE analysis and predictive maintenance system which harvests 400 million data points per month.

Machine alarms, settings, running parameters, and product details, alongside a few additional reliability sensors to collect data – including temperature, vibration and stress wave – are all monitored.

The level of insight now afforded to them is huge. They’re able to see when a machine is not set up optimally, access causal information on why faults occur, and predict equipment failure.

Since implementing this non-invasive change, line performance has increased by 10%.

Choosing the right reliability sensors and using parameters direct from the customer’s own control equipment means no new kit and no factory floor disruption or capital outlay is required.

We’re seeing skilled shop floor operators and supervisors able to target anomaly detection of micro- and macro stoppages, and optimise machine settings such as speed, or component tolerances to drive increased throughputs or reduce waste streams.

To reiterate, AI and machine learning only works with manufacturing and domain expertise. And by working together, we can contextualise data including condition-based parameters, speed, pressure, product, flow, and lubrication timing.

Overlaying this data with reliability data allows us to monitor the health of critical components as well as vital systems such as lubrication pumps where grease levels, system pressure, and temperature can be tracked in real-time to predict potential failures and schedule maintenance proactively.

There really is no limit to the type of components which can be monitored, whether that’s electrical, mechanical or process.

What Does the Future Hold for AI?

The revelation in the Future Factory report that the food and drink sector has been slow to unlock the power of AI and machine learning isn’t surprising.

Some of the biggest players in the industry are accessing as little as one per cent of the data that lies within their own plants.

This means that the FMCG industry is ripe for overhaul; productivity could be transformed but so too could food production.

As the world’s population grows, so does our demand for food. In fact, the amount of food that needs to be produced in the next 35 years is estimated to be more than the total amount of food ever produced in human history.

In truth, the art of manufacturing is rapidly becoming a science, and the AI and machine learning revolution is on its way – whether the sector is ready or not.

By embracing digital transformation, FMCG businesses can create smarter factories that respond dynamically to changing market demands, meet sustainability goals, support growing consumer demand, and safeguard sustainable growth in a rapidly evolving industry. And that certainly provides some food for thought.