January/February 2024 - 21

TECHNOLOGY
TECHNOLOGY
HOLDS POTENTIAL
FOR FOOD
PRODUCTION AND
PROCESSING
By James French
Head of Innovation, TOMRA Fresh Food
W
e've been hearing a lot lately
about artificial intelligence. The
ability of computers to mimic
human thought and decision-making,
to automate tasks that traditionally
required human intelligence, has stirred
up all sorts of sensational news stories.
In confirmation of this technology's
increasing importance, TOMRA Food
has launched two new AI-powered sorting
and grading solutions.
One of those solutions, the TOMRA
Neon, pre-grades machine-harvested
blueberries for the fresh market. The
other, the new-generation Spectrim
X series, leverages deep learning for
unmatched sorting and grading precision
of fruit.
These machines are just the start
of a revolution which will make the
production of fresh and processed foods
more efficient and profitable.
BENEFITS FOR PROCESSORS
The most widespread public discussions
about AI have focused on the latest
version of the chatbot ChatGPT. This
AI-driven language processing tool can
answer questions, perform tasks such as
writing emails and computer code, and
write essays to pass exams that many
students would struggle with. Not so long
ago, none of these AI capabilities existed.
AI also hit the headlines recently
when the technology's godfather, the
cognitive psychologist and computer
scientist Geoffrey Hinton, resigned from
his part-time job at Google so that he
could talk more freely about his concerns
about AI disrupting the job market.
There's no denying that AI-driven
automation will be deployed to tasks
currently handled manually, including
PRODUCE PROCESS ING
21
some of the routine tasks at food
processors and packhouses. As we'll
see in a moment, this will be a positive
development for employers, employees
and food consumers.
TOMRA Food has been utilizing
artificial intelligence since 2019 to
make sorting and grading solutions more
accurate than traditional techniques. The
deployment of AI-powered technology
in the food production industry will
continue to accelerate into the future.
Some of the areas it will impact are:
* Grading and sorting more accurately,
with more stability - resulting in
less food loss.
* Finding difficult-to-detect foreign
materials.
* Improving predictive maintenance.
* Intelligently diagnosing line issues.
* Predicting the useful storage life
of products.
* Optimizing packouts intelligently
for different customers.
* Optimizing supply chain operations.
Make no mistake: AI is bringing a big
technological shift that businesses will
have to keep up with to remain competitive.
FOOD PRODUCTION POTENTIAL
When AI is discussed, you might also
hear the terms machine learning and deep
learning. Machine learning is a collection
of techniques that allows software systems
to recognize patterns in data to provide
measurements and insights. Deep learning
is a subset of machine learning that uses
artificial neural networks to solve complex
problems.
These technologies are well-suited
to food production because many tasks
involve data and decision-making.
AI is also relevant for food production
because of the high level of variability in
the industry, from weather and climate
effects to natural variations in products.
These factors mean that traditional
systems can have a tough time making
accurate predictions. It's not enough just
to have data - its quality is also a big
factor in the performance of AI.
As with traditional systems, the better
the data, the better the decisions. This
is why having the best inspection and
sensor systems matters so much: They
can collect higher quality data, which
supercharges the AI system. This leads
to more accurate and more consistent
decisions which result in less food waste
and more saleable product, as well as
maximizing the product's value.
AI technology can improve sorting
and grading machines in a number of
ways: I can help make more accurate
'accept or reject' decisions; recover more
good product from compromised raw
material through higher accuracy; and
more precisely classify the products on
January/February 2024

Table of Contents for the Digital Edition of January/February 2024
