ŠÓ°ÉAPP

Digital twins: Fonterra Process Optimisation

Initiative type:
Applied research
Sector:
Agriculture and Digital Tech
UniServices Contact:

Opportunity:

Making sustainability improvements at scale in a factory environment is a high-stakes undertaking.

Besides the expense of new equipment, there is the risk that changes will not significantly improve performance or run smoothly. This challenge is amplified when dealing with not just one factory but hundreds, or even thousands, all with slightly different schematics.

With these risks comes the potential for great reward ā€“ from significantly reducing costs and improving efficiency to significantly reducing an organisationā€™s carbon footprint. But rather than relying on trial and error, a calculated risk is much more palatable to large organisations.

Fonterra approached UniServices with exactly this challenge and was hopeful about the potential rewards on the other side.

image across 2 cols
Soft sensor development: On-line viscosity measurement from lab to plant

Modelling the future

Modelling ā€“ creating ā€˜digital twinsā€™ ā€“ was at the heart of how Waipapa Taumata Rau, University of ŠÓ°ÉAPPProfessor Brent Young went about working through Fonterraā€™s challenge. Digital twins are detailed models of factories and processes which mirror their inputs, functions and outputs with a high level of accuracy. Thus, they can predict the effects of certain changes and offer a valuable opportunity to tinker with different variables, with the advantage of being in ā€˜sandboxā€™ mode.

ā€œDigital twins allow you to engage in supported thought experiments,ā€ says Young. ā€œYou can really stress the system and push it in ways you wouldnā€™t push the real process because it would be expensive or dangerous to do. Sometimes you come up with really interesting, counterintuitive results.ā€

Once a digital twin is validated as running realistically, the model follows the laws of physics, chemistry and biology to model what happens when circumstances change. The modelling considers predictable events such as the deterioration of equipment over time, as well as unpredictable ones such as power outages and earthquakes. It can also uncover a myriad of unintended consequences of making certain choices.

right image
Brent Young 400sq
ā€œDigital twins allow you to engage in supported thought experiments. You can really stress the system and push it in ways you wouldnā€™t push the real process because it would be expensive or dangerous to do. Sometimes you come up with really interesting, counterintuitive results.ā€
Brent Young,
University of ŠÓ°ÉAPPProfessor

It can also model changes that are more abstract, such as mapping efficiency if employee numbers were to increase by a certain percentage or what would happen if the factory was organised in a different way. With the help of UniServices, Fonterra was able to take this approach to problem solving and could begin to answer these questions with a couple of clicks of a mouse.

Even aspirational questions can be modelled using digital twins. This is particularly relevant to environmental impact and the production and consumption of energy, or how downsizing could work while maintaining a similar output. Reconsidering traditional ways of working in this way could lead to widespread, positive changes to the way factories operate.

ā€œIf industry doesnā€™t understand a technology or finds it too risky, itā€™s not going to be adopted. Through modelling, we can make sure the technology is understandable and ready for adoption,ā€ says Young.

right image
Could your idea grow into the next great start-up?