We use cookies to offer the best possible user experience on our website. We also use third-party cookies, to deliver personalised advertisement messages. By using our website you agree, that cookies can be saved on your device. Further information on the cookies used and on how to disable them can be found here.
 
 

Artificial Intelligence in VIP's Quality Department "We’re sharing four lessons learned"

The VIP quality department has welcomed a valuable new team member: artificial intelligence. Its purpose? To identify the perfect time for harvesting.

About five weeks before the harvest of a particular apple variety, thousands of apples arrive at VIP's quality department from the various orchards of associated producers. Each apple undergoes testing to determine when those still on the trees will reach the ideal level of ripeness for picking. This is crucial, as Wolfgang Graiss, Head of VIP's Quality Control Department, explains, "for both flavour and the optimal storage that enables us to supply customers consistently all year round."
 
From the human eye to digitalisation…
André Trafoier, an employee in VIP's Quality Control Department, explains that one of the parameters for assessing an apple's ripeness is its starch content. To measure this, apples are sliced and sprayed with an iodine-potassium solution, to which they react: the lighter the slices remain, the higher the starch content and the less ripe the apple.
Until a few years ago, this observation was carried out by the naked eye, with values calculated manually. Over time, the process was digitalised: now, a software calculates average values, while a camera captures and saves photos of each slice, ensuring everything is traceable and well-documented.

… from digitalisation to artificial intelligence
Since October 2023, VIP has been testing AI, comparing starch levels as assessed by employees with those suggested by the AI. To "train" the AI, over 27,000 apples of various varieties and maturity levels were photographed in the months leading up to its first use. The accuracy of the AI's analysis depends on the quality of the data with which it is "trained".
Trafoier notes: "The results were excellent for varieties and ripeness levels with extensive data available”.

Ready to share some key insights on AI
From this experience, VIP has gained valuable insights into AI and is keen to share them with other companies considering similar projects.
  • First, it’s essential to ask whether true AI is necessary or if an intelligent digitalisation of the process will suffice.
  • Second, training an AI requires gathering a large volume of data.
  • Third, the project should be broken down into small steps so that goals are achievable one step at a time, reducing business risk.
  • Finally, employee acceptance is crucial: involving the team from the earliest stages of the project strengthens their motivation.

A local artificial intelligence
It’s worth noting that VIP, after carefully considering proposals from software developers in New Zealand and Canada, found the ideal partner close to home, in Naturno, just 12 kilometres from VIP’s headquarters in Laces, where its main cooperatives are based.
"Choosing to work with Naturno-based IT company Fuxware" remarks Andreas Oberhofer, Head of VIP's IT Department, "turned out to be the right decision from the start. They worked incredibly quickly and are always on hand to answer our questions".
 

Share

True friends share everything

More articles

Also interesting

 
All articles