The precision data driving sustainability gains in dairy processing

12. juni 2024 | FOSS

• Food Lab
• Food Processing

downloadKun til redaktionelt brug

Automated process control based on key data from the production line is helping to improve efficiency and sustainability across Arla production plants. Process consultant René Klinge Iwang gave us an update on the concept and explains how the reliability of sensors providing the data is a critical factor in any control system.

Never mind the steel pipes of the dairy plant, it also takes nerves of steel to balance high-volume processes against wafer thin profit margins and the universal imperative to make the very best use of resources.

In recent years, that level of control is challenged by competition for dairy engineers, especially in the home of country of the international Arla cooperative where a booming Danish pharmaceutical industry is often successful in tempting talent away from the unglamorous fluorescent-lit control rooms of dairy processing plants. Young dairy professionals would nonetheless be well-advised to take a closer look at what they are missing out on because it is in dairy engineering that tangible progress is being made towards the goals of industry 4.0. And it is happening right now, driven by the desire to deliver nutritious products to an increasingly demanding consumer market while constantly improving the use of valuable dairy resources.

“Today, it is all about data,” says Arla process consultant, René Klinge Iwang. “Data is the key to everything because we are so much focused on the whole industry 4.0 idea. So every decision we make has to be based on good data.”

Automate to save
This informed decision-making approach is based largely around inline sensors transmitting from the heart of established process control systems. Good progress is being made with many different key performance indicators in place in many different processes. “We have to make sure that we get the maximum out of every kilo of fat and protein. By making sure that we get it, we also make sure that it's sustainable as possible and that's why we are constantly monitoring all of our processes,” says René.

Just one example of how this control pays off is in the running of energy-intensive processes such as pasteurization and ultrafiltration. Process control helps to monitor how much energy is being used in different steps. It is also possible to monitor when it's time to replace membranes and such like that are prone to wear. Similarly, tight process control is paying off when it comes to using key parameters to match product quality goals while also making a profit from the process.

Walking the profit tightrope
Having precise data on parameters such as moisture in butter, protein in fresh cheese, the fat to protein ratio in cheese milk and many more, enable closer control according to predefined production targets. This helps to optimize the use of raw material and reduce the giveaway of expensive components. While much can be done with manual control, it is when automated systems working on a continuous stream of data from the process get to work, that the results really start to show. “For at least some of the processes, mainly the powder drying towers, we are actually using what we call APC, so fully automatic process control,” René explains.

He describes how the gains afforded by automated control systems are most apparent when navigating the super fine margins of today’s dairy production. For instance, one of the production lines he has worked with produces around 75,000 tons of cream cheese per year. At this volume, small adjustments to fat content of 0.1% can have a huge effect on profit without affecting pre-defined quality standards. Then there are the follow-up gains working with the permeate. “Whatever the fat per kilo price is, it's crazy how much value is at stake – it goes so fast,” says René.

The cows and the grass
When working with such large volumes and such tight margins, the whole control system has to work perfectly day in, day out and this naturally places a lot of weight on the data it is running on and the sensors providing that data. “We need the answers to show the right value all the time to the operator so they can adjust that process,” says René.
For parameters such as moisture, fat, protein and others, the sensor in question is invariably an analytical instrument specially designed to work in the process environment. But having the right analytical hardware is one thing, the other is to ensure that it is always correctly calibrated. “I don't think it is so much the technology that is the limiting factor. I think it's more the calibration adjustment and modeling,” says René. “We check and adjust calibrations according to things like the changes in milk quality that occur when cows move onto grass in the spring. And that variation you will very much like to have into the prediction model. It's so important that the calibration model includes some samples that are from the real production. You can always have some standard prediction model with some generic data from all over the world, but the model is so much better if you have concrete data from the production.”

Building on solid ground
Thinking about all the equipment and all the data to be managed, it could sound like a whole lot of work to run today’s advanced process control systems, defeating the very principle of reducing the burden on individuals running today’s dairy production plants. But once again, technology comes to the rescue. For instance, using the latest networking and connectivity technology, a system has been set up to monitor all calibration models on all instruments so that the data generated is constantly evaluated against reference laboratory results.

It's an approach that promises to raise standards by helping a small workforce to manage process control throughout the organization. “We have 60 different sites and you know, we don't always do things the same way at every site,” says René Klinge Iwang. “So this is a big challenge, always to adapt, to look at what we are doing today and what can we do better tomorrow.”