A Cornell-linked startup is testing technology that could help Finger Lakes apple growers sort fruit at harvest rather than wait for packing-house inspection. Ficio, founded by Cornell engineering graduate student Youhanna Meleka, is working with farms in Ontario and Wayne counties as it develops the system.
The company's FieldSorter uses computer vision to identify apples with defects, including bruises, punctures and clipped stems, while moving through orchard rows, according to a Cornell Chronicle report. It records a grade for each apple and links that information to the tree where it grew.
That approach is intended to give growers earlier information about which harvested apples can be sold fresh and which should go to other uses. Ficio says growers now can wait until fruit reaches a packing house, sometimes after months in controlled storage, before getting that determination. The company has not released measured savings from its system, and the technology remains in development.
Ficio demonstrated FieldSorter earlier this year at Cornell AgriTech in Geneva before about 300 growers. It is working with farms in Ontario and Wayne counties on further field pilots and adjusting the equipment based on growers' feedback. The Cornell report did not identify the participating farms or announce a commercial launch date.
āGrowers spend a lot of time and money finding out which apples are market-ready and which arenāt,ā Meleka told Cornell. āWeāre helping them make that determination at harvest rather than months later in a packing house.ā
One platform for farm and kitchen
The startup is also developing FicioPrep, a robotic station for large kitchens. The prototype uses computer vision to recognize ingredients and assess potatoes before cutting them into fries or chips. Meleka demonstrated it to Cornell Dining managers as the company gathered feedback, but Cornell did not say the dining operation has adopted it.
Meleka said he hopes FicioPrep eventually can handle more vegetables, prepare ingredients on demand, clean itself internally and record processing volumes for kitchen managers. Those are development goals, not verified capabilities in current commercial use.
Both products run on a software platform Ficio calls FoodMind. The company says it is designed to work on different kinds of machines, allowing lessons from one task or environment to inform another. Its longer-term aim is to develop machines that pick, sort, pack and prepare food across the supply chain.
Cornell support and next tests
Meleka, who grew up in Cairo, Egypt, said his interest in the project comes from seeing food scarcity and waste. He studied logistics and manufacturing at Penn State and worked as a global supply manager at Tesla before enrolling in Cornell's Master of Engineering pathway in semiconductor and quantum materials.
At Cornell, he has worked with mentors through Entrepreneurship at Cornell and the Rev: Ithaca Startup Works incubator, drawing on commercialization, manufacturing and business-development advice. Ficio also works with industry partner Universal Robots and belongs to NVIDIA's Inception program, which provides developer tools and training.
Zach Shulman, director of Entrepreneurship at Cornell, said the apple-sorting system has the potential to save growers resources. Its effect on labor, fruit losses or farm income will depend on how the upcoming field pilots perform; the report supplied no independent trial results or cost figures.
Meleka said early plastic and wood prototypes were taken to farms so growers could describe what would be needed in real orchards. The current work in the two Finger Lakes counties is another step in that testing process, rather than evidence that the system is ready for widespread deployment.
FingerLakes1.com is the region’s leading all-digital news publication. The company was founded in 1998 and has been keeping residents informed for more than two decades. Have a lead? Send it to news@fingerlakes1.com.
