On the way to the intelligent food factory
Healthy and as convenient to consume as possible – these are the “trendy” foods. The demand for them is growing. Many manufacturers are investing in additional production capacity. But everyone else is also required to further improve the efficiency of their production. Those who modernise and do not shy away from innovative technologies such as AI will remain competitive.
There is increasing talk of the “smart factory”. For a long time, this only existed in the manufacturing industry. The prototypes were car factories, equipped with all kinds of data collection tools, Wi-Fi and 5G networks, ready for big data analyses, with the digital twin in the background.
What does this have to do with food production? A lot – and more all the time. The “smart food factory” is no longer just a vision. With solutions offered by the Industrial Internet of Things (IIoT), it is now easier to meet fluctuations in demand, the requirement for greater delivery frequency and the high demand for safe, high-quality food. Once the first step, the extensive automation of production, supported by numerous sensors, actuators, and quality analytics, has been completed, the next steps towards the smart food factory are imminent.
Predictive maintenance as an intermediate step
According to the DLG Trend Monitor 2024, 42 per cent of companies are planning to invest in digitalisation within the next three years. 39 per cent want to further increase automation. 52 and 50 per cent of companies respectively will invest in replacement investments and increasing efficiency. A total of 36 per cent plan to invest in capacity expansions. The trend monitor also shows that modern maintenance methods are used by 29 per cent and are planned by 26 per cent. So far, 10 per cent of respondents have used AI; 34 per cent plan to use it in the next three years. Nevertheless, most are still a long way from the smart factory. At least ten per cent already refer to their factory as such. A further 23 per cent want to realise it soon.
The digital transformation is also finding its way into medium-sized food producers such as Kräuter Mix: with networked systems and the visualisation and optimisation of production processes.It’s not just the big players that are using AI
In July 2024, for example, Danone announced that it was looking into a collaboration with Microsoft. The aim is to integrate artificial intelligence (AI) into all operational processes. The Group has been driving forward the use of digital technologies for several years now. The lighthouse factory in Opole, Poland, plays a central role in this. Unilever uses AI in production to increase efficiency and reduce waste. Kraft Heinz uses AI to create demand forecasts and attempts to speed up product development. More and more medium-sized companies are also discovering the benefits of AI in production: for example Kräuter Mix (Germany) – for monitoring the production process and forecasting demand.
“Digital food safety” is one of the key topics. Forecasting tools make it possible to quickly deduce the effects of changes in raw materials or production on product safety. Nestlé is currently pushing ahead with this. The Nestlé Institute of Food Safety and Analytical Sciences (NIFSAS) in Lausanne and other institutes in China and India play a leading role in this. Among other things, the teams are developing robust analysis methods and tools for artificial intelligence. In future, big data will be used in real time to identify risks.
Broad range of intelligent sensor technology and PAT
For those who cannot rely on self-developed methods, there is a growing range of products on the market. More and more sensors are meeting the hygienic requirements of the food industry as well as the specific features of certain production areas. For example, level measurement can be realised in the event of heavy dust formation, such as in grain silos, as well as in small storage containers for baking ingredients. Cloud-based fill level sensors enable remote inventory management. The use of radar-based moisture sensors is aimed at high product quality and reduced energy consumption. Digital sensor technology that stores a large amount of data offers future security. This supports predictive maintenance and IIoT services.
Process analysis technologies such as Raman spectroscopy for determining the chemical composition, pH, concentration and density measurement, conductivity, colour, and turbidity measurement are suitable for the inline quality control of foodstuffs. The wide range of applications not only guarantees product quality, but also increases plant availability. CIP cleaning can also be optimised through inline concentration measurement and determination of the coating thickness at critical points in the system in conjunction with level and flow measurement. Instead of time-controlled processes, the cleaning process is monitored and controlled in real time. This reduces cleaning times and ensures traceability.
Communication via IO-Link connection technology also enables factory automation, including extensive diagnostics. Sensors with hygiene adapters are also a good choice for hygiene-sensitive processes.Ethernet right into the field
In addition to intelligent sensors for data collection, the next steps towards IIoT and the smart factory require powerful, end-to-end networking. It must transmit large volumes of data at high speed from the field level. In non-Ex areas, which is the norm in the food and pharmaceutical industries, Single Pair Ethernet (SPE) is suitable for this. It uses a two-core cable to transmit both data and power. Ethernet-APL is also available for hazardous areas.
Extensive automation as the basis of the smart factory
But no manufacturer should take the second or third step before the first. If all ingredients are automatically weighed or dosed according to the recipe, if the subsequent process is precisely controlled and the entire production process, including any manual interventions, is managed automatically – only then can the transition to the smart factory succeed. But even without this requirement, plant operators benefit from a high degree of automation: for example, through extensive production transparency and easier batch tracing. Product safety increases. Linking production and maintenance systems generally increases plant availability. Anyone who then dares to connect the production system to higher-level MES and/or merchandise management systems is closer to optimised operations management even without AI – and therefore certainly competitive.
