6 min read • published in partnership with ITI Group
Insight: How food and drink manufacturers can avoid islands of automation
Automation is no longer optional for food and drink manufacturers under pressure to improve productivity and manage labour shortages. But without a clear approach, investment creates fragmented systems that deliver limited gains. ITI Group’s Oliver Stone explains why success depends on how well automation is connected, applied, and scaled.
Food and drink manufacturers are under pressure from every direction. Margins are tight and getting tighter, squeezed by retailer pricing power and rising input costs. At the same time, roles are becoming harder to fill and retain.
Workers are often required to carry out physically demanding, repetitive tasks on high-throughput lines operating 24/7. For years, labour gaps were managed through agency work and low-cost immigration, but that model is under strain. Vacancy rates* in food and drink sit at 5%, more than double the manufacturing (2.1%) and national (2.3%) averages.
It’s no surprise that automation has moved up boardroom agendas. Around one in four food and drink manufacturers (23%) are now looking to automation to address labour shortages, jumping to 40% among SMEs.
For many, this is their first serious investment in automation. The good news is it’s more accessible than ever. Robotics, collaborative robots (cobots), automated guided vehicles (AGVs) and intelligent connected machines are now within reach of even micro-businesses.
But as manufacturers rush to automate, we see the same mistake being repeated. There’s a reason almost three in four automation and digital initiatives fail to meet their intended goals, and it’s not because the technology doesn’t work.
The trap that catches manufacturers
The assumption too often made is that more automation leads to better performance. What usually happens instead is that businesses create islands of automation that don’t communicate, can’t be managed centrally and still relies heavily on people to function.
We regularly see manufacturers invest significant capex into equipment that improves a local KPI without improving overall line performance. A high-speed packing robot installed downstream of a slower manual process. A vision inspection system capturing quality data that never feeds back to the process actually causing the defects. Autonomous warehouse robots operating efficiently within their own zone but completely disconnected from production scheduling.
Each element works, but collectively the system doesn’t. In most cases, efficiency remains flat or even drops because complexity has increased, points of failure have multiplied and constraints have simply shifted elsewhere.
Automation on its own doesn’t guarantee improvement. It must be part of a coordinated, system-level approach where each investment contributes to a more efficient, more responsive whole. The path to getting there follows five steps.
Step 1: Start with the constraint
Most automation conversations start with what the technology can do. The better ones start by asking what’s actually limiting performance. That might be a bottleneck limiting throughput, recurring downtime on a critical machine, labour volatility in a high-churn process, or yield loss in a specific production stage.
The key is specificity. If the constraint is tightly defined, the solution will be too – and so will the value it delivers. A manufacturer we worked with was investing in end-of-line automation while their real constraint was an unpredictable changeover process upstream. Fixing the right problem first unlocked performance improvements that had previously seemed out of reach.
Step 2: Think big, be targeted
We have a saying: be ambitious enough to drive genuine change; pragmatic enough not to over-engineer the solution. Your end goal should absolutely be ambitious, but trying to get there in one leap is where programmes tend to stall.
The approach that works is to focus on a specific area, deploy a contained solution and prove the impact before moving on. Every successful programme we see follows this approach. One well-executed step at a time builds momentum, creates internal confidence and helps teams learn what works in their own operation. Remember also, that this is a journey, one that can’t be rushed. Iterative, long-term change delivers far more than trying to solve everything at once.
Step 3: Build the digital layer without overcomplicating it
As improvements are deployed, data starts to flow across the operation. This is often where things become more complex than expected. Different machines speak in different languages. Data is inconsistent and frequently lacks context (time, batch, product, condition). Without standardisation and context, data isn’t all that useful.
We’ve seen sites generating plenty of data but drawing very little insight from it. The priority at this stage isn’t more data, but usable data: aligned to production events, consistent across systems and accessible at line and site level. This becomes the foundation for everything that follows.
Step 4: From visibility to control
Once data is reliable and connected, the focus shifts from visibility to orchestration. This is where things get really exciting. Rather than relying on operators to interpret dashboards and reacting manually, connected systems can begin to detect issues earlier – such as drift in quality parameters – trigger predefined responses like adjusting machine settings or rerouting flow, and coordinate actions across the line in response to production demand.
This is a meaningful shift. Operators move from firefighting to oversight, problems are resolved before they become stoppages, and the operation becomes measurably more consistent.
Step 5: Building towards lights-out manufacturing
Fully autonomous factories are still rare in the UK. What we are seeing, however, is a steady shift towards partial autonomy: equipment that flags its own maintenance needs, quality systems that self-adjust within defined limits, and production plans that dynamically adapt to demand and constraints. Each step reduces the need for manual intervention, improves consistency and makes the operation more predictable.
Fazer, Finland’s leading confectionery manufacturer, is a good example of this journey in practice. After ITI supported a trial of connected digital technologies in their existing facility and seeing tangible improvements, the decision was made to embed the full approach into Fazer’s new €400m chocolate factory, currently under construction.
Rather than retrofitting an existing operation, ITI has been involved in designing the factory from the ground up: automated lines that communicate directly with the digital layer, building intelligence over time and enabling the facility to orchestrate itself. It will operate with a fraction of the workforce of its predecessor. That is what a methodical, step-by-step approach to automation ultimately makes possible.
The operational backbone of a connected factory
Connecting these layers requires a system that can coordinate execution across the operation. This is where manufacturing execution systems (MES) play a central role. When deployed correctly, it links production, materials management, quality control and scheduling – providing the structure needed to turn data into action, and freeing production managers from chasing information to focus on orchestrating operations.
MES has historically been seen as complex and expensive. That perception is changing. Modern platforms are modular, cloud-based, fast to deploy and designed for phased adoption. Manufacturers can start small – basic performance tracking on a single line – and scale as value is proven.
Turning strategy into action
Across the UK food and drink sector, most manufacturers are at step two or moving into step three. They’ve introduced automation in parts of the process, but aren’t yet seeing the full benefit because systems aren’t properly connected.
At ITI Group, we help manufacturers define a path forward – identifying where to start, avoiding common pitfalls, and ensuring each step delivers measurable value today while building towards something more connected and resilient over time.
A good first step is our free Digital Readiness Assessment, a practical tool that helps define priorities, highlight risks and shape an improvement plan based on how your operation actually runs – not as it looks on paper.
The pressures facing food and drink aren’t going away. The question isn’t whether to transform, but how to do it in a way that compounds over time. That’s what a connected, step-by-step approach makes possible.
Visit ITI Group at Smart Manufacturing Week, NEC, 2nd and 3rd June, stand F144, to discover how we’re driving smarter, more efficient manufacturing solutions.
