Cancel OK

Sunkist machine uses AI to sort citrus

sunsort citrus sorter

Sunkist BB #:102141 has brought online a new machine that uses artificial intelligence (AI) to sort 12 pieces of citrus fruit per second per lane.

It’s called Sunsort, and it uses Google AI technology “to automatically identify fruit characteristics such as size, shape, weight, blemishes, and even the decay that might not be visible to the naked eye,” says a company press release.

Sunsort returns data to the operator, who decides where he or she wants to send the fruit, says José Flores, software and electrical engineering manager at Sunkist Research and Technical Services (RTS), the research arm of the citrus giant.

“Bringing AI to the fruit sorting process allows our packhouses to dramatically increase the speed, accuracy, and efficiency of their sorting operations,” says RTS general manager Aaron Gorsky. “Sunsort equipment can conduct a comprehensive fruit assessment and sort the piece accordingly, all in one pass. This offers substantial operational efficiencies and improvements.”

The operator manages the sorter’s settings, which dictate which bin each piece of fruit will go to. These settings can be changed manually.

This flexibility is important because “the market changes: what might be first-grade at the beginning of a season might not be first-grade at the end,” notes Flores.

With Sunsort, the fruit is “running right into boxes,” Flores adds. These then go to a pallet without any human contact.

“One of this AI sorter’s greatest features is its ability to learn,” explains Gorsky. “Through repetition, the sorter learns to distinguish a defect’s appearance from normal anomalous features like blossoms and stems. This happens in one pass, so it eliminates the need for special lines, manned by separate sorting specialists, who focus on specific blemishes. It’s a huge time saver.”

Sunsort has been in development for three years and has been rolled out in five Sunkist facilities in California over the past year. It is also used in four facilities in Argentina and five in Spain, which are not owned by Sunkist.

After its success in Sunkist packinghouses, Sunsort is now available “to any processing facility that needs to increase revenues through fast and accurate sorting,” says the company.

“By upgrading to our latest technology, Sunsort ai, you can significantly reduce the manual grading labor cost,” says Mina Abdelshahid, RTS director of operations. “Our customers are seeing returns on their investments within only one and half years from labor saving alone. Plus, Sunsort’s increased processing power allows you to increase the sort’s speed by 10 percent, making your overall operation more efficient.”

Twitter

Sunkist BB #:102141 has brought online a new machine that uses artificial intelligence (AI) to sort 12 pieces of citrus fruit per second per lane.

It’s called Sunsort, and it uses Google AI technology “to automatically identify fruit characteristics such as size, shape, weight, blemishes, and even the decay that might not be visible to the naked eye,” says a company press release.

Sunsort returns data to the operator, who decides where he or she wants to send the fruit, says José Flores, software and electrical engineering manager at Sunkist Research and Technical Services (RTS), the research arm of the citrus giant.

“Bringing AI to the fruit sorting process allows our packhouses to dramatically increase the speed, accuracy, and efficiency of their sorting operations,” says RTS general manager Aaron Gorsky. “Sunsort equipment can conduct a comprehensive fruit assessment and sort the piece accordingly, all in one pass. This offers substantial operational efficiencies and improvements.”

The operator manages the sorter’s settings, which dictate which bin each piece of fruit will go to. These settings can be changed manually.

This flexibility is important because “the market changes: what might be first-grade at the beginning of a season might not be first-grade at the end,” notes Flores.

With Sunsort, the fruit is “running right into boxes,” Flores adds. These then go to a pallet without any human contact.

“One of this AI sorter’s greatest features is its ability to learn,” explains Gorsky. “Through repetition, the sorter learns to distinguish a defect’s appearance from normal anomalous features like blossoms and stems. This happens in one pass, so it eliminates the need for special lines, manned by separate sorting specialists, who focus on specific blemishes. It’s a huge time saver.”

Sunsort has been in development for three years and has been rolled out in five Sunkist facilities in California over the past year. It is also used in four facilities in Argentina and five in Spain, which are not owned by Sunkist.

After its success in Sunkist packinghouses, Sunsort is now available “to any processing facility that needs to increase revenues through fast and accurate sorting,” says the company.

“By upgrading to our latest technology, Sunsort ai, you can significantly reduce the manual grading labor cost,” says Mina Abdelshahid, RTS director of operations. “Our customers are seeing returns on their investments within only one and half years from labor saving alone. Plus, Sunsort’s increased processing power allows you to increase the sort’s speed by 10 percent, making your overall operation more efficient.”

Twitter

Richard Smoley, contributing editor for Blue Book Services, Inc., has more than 40 years of experience in magazine writing and editing, and is the former managing editor of California Farmer magazine. A graduate of Harvard and Oxford universities, he has published 12 books.