In an era of rising tariffs and declining margins, operational efficiency isn’t just a requirement – it’s an absolute necessity. The food manufacturing industry is no exception and, given the numerous challenges facing it, requires innovative solutions that enhance efficiency in the face of increasing competition, tightening regulations and shifting consumer preferences. That means manufacturers not only streamlining production but also ensuring sustainability and quality. Self-evidently, increasing efficiency in the production process necessitates the use of ever more innovative machinery. From his perspective, David Dons, director of digital & optimised production at Arla Foods argues: “The first step to increase efficiency is to create technology standards per equipment category, so that we can have the same approach across all Arla Foods sites and countries.
“The second step [for us] is to continuously develop the standard and partnership with equipment suppliers to ensure Arla has the best available technology and to take part in future developments to improve efficiency. At the same time – and with efforts across the Arla Foods organisation – we explore new and innovative products, processes and technologies to develop enhanced ways of producing our nutritious dairy products so as to increase yield and quality, while improving efficiency.”
Or to put it another way, future trends in food production efficiency point to increased reliance on technology and sustainability. Leading the way will be advancements in AI and robotics, which will automate more processes from farm to table, for example, thereby ensuring precision and consistency.
Unsurprisingly, Dons agrees: “At Arla, we believe the future of food production efficiency lies in leveraging advances in technology and the technical innovation of our teams to unlock the highest possible value from every drop of our farmer-owners’ milk.
Our responsibility is twofold – to create nutritious products that will continue to feed the growing populations in Europe and beyond, and to do so while lowering our climate footprint across our value chain from farm to fridge.”
Maintaining dialogue with industry experts
Advances in digitisation and automation tools are already helping the company to reduce waste, optimise its energy utilisation and ensure consistent product quality, according to Dons. “Beyond exploring the new opportunities technology presents, we also remain in close dialogue with industry experts, our farmer-owners and our customers, all while collaborating actively with universities and local organisations to discover how we can push efficiency and sustainability even further across our operations,” he says.
Yet what makes the difference at Arla, insofar as Dons is concerned, is its people. “We are fortunate to have colleagues who are passionate about continuous improvement and innovation and their ideas, which continuously spark new improvements. Their skills and experiences combined with new, sustainable and ever-advancing technologies, are what will enable us to continue meeting rising global demands while ensuring that every product we create is both high in quality and responsibly produced.”
Optical sorting – the way forward
In general terms, one major takeaway emerging from the food production process in recent years has been the depth and scale of innovation – one significant example being optical sorting. This is an automated production process using cameras, lasers and (now) machine learning to identify and remove defective or foreign materials from food products. The result is not only higher quality and safer food for consumers, but also improved efficiency by increasing sorting speed and minimising human error.
Market research firm Fortune Business Insights valued the global optical-sorter market at $2.45bn in 2024 and projects growth to $4.32bn by 2032 (CAGR 7.4%) Other analysts give different estimates, with 2023–24 market values generally falling in the $2.1–3.1bn range. Growth in this market has been driven in part by increasing demand for automation in various industries such as food and beverage, pharmaceuticals, recycling and mining with sorting algorithms becoming ever more sophisticated.
Renewed emphasis on sustainability
Meanwhile, the recycling industry has been leading the way in response to heightened environmental concerns and the renewed emphasis on sustainability.
As the European Commission’s Packaging and Packaging Waste Directive (PPWD – Directive 94/62/ EC) highlights, the use of optical sorters has practical implications for various industries.
Packaging is defined as “products made of any materials of any nature used to contain, protect, handle, deliver and present goods, from raw materials to processed goods, from producer to user or consumer”.
By 31 December 2025, at least 65% of packaging waste must be recycled, calculated by weight. The recycling targets per material are: 50% of plastic; 25% of wood; 70% of ferrous metals, 50% of aluminium; 70% of glass; and 75% of paper and cardboard. By 31 December 2030, at least 70% of packaging waste must be recycled by weight, including 55% of plastic. In short, the expansion of collection systems to include all flexible plastic packaging formats will be a likely requirement in all EU member states in order to achieve these ambitious targets. Unsurprisingly, the collection of flexible plastic packaging commingled with other plastic packaging, metal containers and drink/liquid cartons in the lightweight packaging category is the preferred industry approach to maximising sorting and recovery efficiencies.
That these objectives are achievable is principally due to state-of-the-art sorting facilities for lightweight packaging and flexible plastic packaging that can be effectively captured using wind-sifters and ballistic sorters to aggregate flexible plastic packaging. It can then be sorted by optical sorting units that are specifically adapted to manage flexible plastic packaging and films. Hence, these machines can also be used to efficiently sort and separate other recyclable materials such as glass and metals. The same principle applies in the food manufacturing space where sorters similarly use optics to analyse products that need segregation.
While there are significant opportunities across various food types, with fruits and vegetables holding a major share, followed by grains and nuts, the key point to consider is that technological advancements in sensor technology – such as the use of software algorithms and machine learning – are raising both the accuracy and speed of optical sorting.
Though the initial capital investment required to install these machines may be substantial, the longerterm numbers stack up in the form of reduced wastage and improved product quality. Using cameras and lasers to determine the colour, shape, size and structural properties of food items, the optical sorter typically should be able to spot subtle defects such as discolouration, as well as more obvious issues such as stones, plastic and glass.
Detection on a sub-millimetre scale
Items identified as defective or containing foreign material are removed from the production line by mechanical arms, air jets or conveyor belt redirection, for example. Early optical sorters, according to Hamid Abdullayev at the Azerbaijan State University of Economics, used single-point photoelectric sensors that could only detect the presence or absence of light (as a basic light/dark threshold).
“These have now been replaced by high-resolution CMOS sensors in camera arrays, allowing for detailed imaging over a wide area,” he says, adding: “Modern sensors can detect defects on the sub-millimetre scale. This is crucial for premium products where even small defects can affect their marketability.”
This has been aided by the dramatic improvement in LED lighting, providing consistent, controlled illumination across multiple wavelengths (UV, IR, visible). This enhances detection of otherwise hardto- see defects; in some applications hyperspectral/ fluorescence imaging can flag kernels likely to carry mycotoxin contamination, although laboratory testing is needed to confirm levels.
Higher throughput, reduced waste
Collectively, these improvements allow for higher throughput, reduced waste and better quality control, which is essential for commercial operations that sort many tonnes per hour.
“Over the past few years, optical sorting machines have undergone significant advances in speed, accuracy and intelligence, allowing them to handle large volumes with high precision,” Abdullayev adds.
“Today’s machines use line or area scan cameras with megapixel resolution and high frame rates (often over 10,000 lines per second). These allow for the detection of finer defects (e.g. insect damage, micro discolouration) even at high belt speeds.”
Sensors are no longer passive; they are integrated with external computing systems that make instant decisions and adapt in real time. Some systems now use self-calibration and machine learning to reduce set-up time and operator intervention. Some of the latest machines combine multiple sensor modalities (e.g. visual + NIR + X-ray) into a single unit. This combination provides a more complete image of the object, increases accuracy and reduces false rejections.
In short, combined with appropriate filters and lighting, this allows machines to detect elements invisible to the human eye, such as moisture, internal bruising, or fungal contamination. It also allows for improved differentiation between good and defective products based on chemical composition, not just colour or shape. If sensors are both the eyes and brains of modern sorters, then their increased accuracy, speed and intelligence are directly responsible for improving both the efficiency and quality of food sorting processes, according to Abdullayev.
Modern sorters are increasingly deploying AI models (particularly convolutional neural networks) to learn from large datasets and classify products more effectively. This helps reduce false positives/negatives and increases the ability to adapt to new product types or defect profiles.
Clearly, for consumers, the risk of contamination diminishes as do the chances of contracting a foodborne illness. For the manufacturer, meanwhile, efficiency is improved because sorting is conducted at a much faster pace than would ordinarily be the case if done manually.
For consumers and producers, both will likely benefit in the long term.
