Gentle de-nesting and tray/pack loading of fragile cookies, crackers and wafers
Across its large bakery operations (cookies, crackers, wafers, cones from the Weston Foods acquisition), the company must transfer thin, brittle baked pieces from cooling/conveyor flow into trays, thermoform cavities, or pack patterns for inspection and packaging. The pieces are crushable, vary in thickness and curvature batch-to-batch, and chip or snap under even modest point loads or pinch grip. This task sits downstream of baking/cooling and upstream of flow-wrapping and cartoning. It is hard for a robot because the grip force must be tuned per piece to lift without fracturing while still defeating slip, and because variable, fragile geometry makes a vision-only pick prone to crushing or dropping. It is a high-throughput, repetitive handling step today done by line workers, but the contact-sensitivity is what distinguishes it from robust bulk pick-and-place. We identified this through our own research; we have not confirmed the specifics with the customer directly. This page is our researched read — a starting point for that conversation.
What the task is
RESEARCHED · our reconstructionAcross its large bakery operations (cookies, crackers, wafers, cones from the Weston Foods acquisition), the company must transfer thin, brittle baked pieces from cooling/conveyor flow into trays, thermoform cavities, or pack patterns for inspection and packaging. The pieces are crushable, vary in thickness and curvature batch-to-batch, and chip or snap under even modest point loads or pinch grip. This task sits downstream of baking/cooling and upstream of flow-wrapping and cartoning. It is hard for a robot because the grip force must be tuned per piece to lift without fracturing while still defeating slip, and because variable, fragile geometry makes a vision-only pick prone to crushing or dropping. It is a high-throughput, repetitive handling step today done by line workers, but the contact-sensitivity is what distinguishes it from robust bulk pick-and-place.
Is this the actual task and sequence? What are the real tolerances, cycle rate, and reject criteria, and which steps are today's manual bottleneck? Answering these is what turns this from a researched signal into a validated use case.