Hard numbers always trump soft benefits when judging automation projects, but the case for soft paybacks is getting stronger and more compelling for food companies coming to grips with higher food safety requirements and labor-availability issues.
Those calculations can’t be made, however, unless automation can reasonably duplicate human movement. That’s one of the major challenges for robotic systems, particularly as they move further upstream in packaging and processing applications. End-of-arm (EOA) tools must be able to handle products with the same dexterity and gentleness of the human hand.
Vacuum often is involved when fabricating such tools, but an alternative approach is being applied in a pilot project at Taylor Farms Pacific, Tracy, Calif. The early-stage project involving EOA tools fabricated by Soft Robotics Inc. is under way to see if a pick-and-place robot can effectively assemble salad kits and transport easily damaged grapes and other produce.
The tool’s grippers are inflated to 1-13 psi, depending on the rigidity required, and can run through four inflation/deflation cycles every second, according to Carl Vause, CEO of Cambridge, Mass.-based Soft Robotics (www.softroboticsinc.com). Fabricated from surgical-grade Delrin acetyl resin, the grippers were inspired by the tentacles of an octopus and can handle payloads of 200g/7 oz. to 2kg/4.4 lb. Grippers that can handle 13-lb. loads are being tested. The technology is licensed from research chemists at Harvard University.
Bakery, poultry and produce represent the most likely opportunities for the robotic work cells, Vause believes. Sanitary design is reflected in the materials of construction. Soft Robotics expects to receive certification under new robotic sanitary standards developed by 3A, which could pave the way for use in dairies and other hygienic applications.
When interfacing with robotic arms, the tools are agnostic. Seven-axis articulated arm robots from Yasakawa Motoman were integrated with them by Henzen Manufacturing International, and ABB Flex-Pickers were used by JLS Automation for the Taylor Farms test. “It’s no longer a project, it’s a product that people already are using,” says Vause.
In a testimonial, Taylor Farms president Alan Applonie calls the pilot “the most exciting project I’ve ever worked on in my career.” Elimination of ergonomic injuries and improved accuracy and precision in product placement are benefits, but adaptability may be the biggest advantage. The short lifecycle of most products often thwarts payback from automation projects, he notes; the adaptability of the EOA tool increases the likelihood of successful redeployment if a product flops.
Pick, inspect & place
While Taylor Farms hopes to harvest a bumper crop of soft benefits, it already is reaping a hard benefit: system cost. Most patents for Delta robots have expired, resulting in plunging prices. More than 50 Delta OEMs now fabricate those pick-and-place arms, and prices are half what they once were, according to some system integrators.
New suppliers are offering manufacturers more than lower prices. Added benefits include more axes of motion, longer reach and heavier payloads, with a Delta arm capable of handling 90kg (198 lbs.) available.
Enhanced functionality that adds payback value is another change, with 3D vision extending the machines into quality-assurance tasks. Besides guiding pick arms, two-dimensional cameras have helped to reject products with imperfections like burn marks on buns for some time, notes Craig Souser, president of JLS Automation (www.jlsautomation.com), York, Pa. But 3D represents “a new dimension in capabilities,” allowing automated systems to evaluate product attributes such as voids in energy bars or the height of stacks of sliced deli meats.
Faster algorithms and advanced imaging systems make inspection possible without slowing down the robot. JLS is integrating into a system a Cognex In-Sight camera with a dedicated processor. “That provides plenty of bandwidth to do the calculations,” says Souser. He notes the camera also performs tracking and guidance duties, functions typically performed by a 2D camera.
Hygienic design is one of food processors’ top considerations when automating processes, and Souser doesn’t believe the current generation of collaborative robots measures up to industry sanitary standards. Downstream processes are less demanding, however, and collaborative machines are entering the market for secondary packaging activities, such as palletizing.
Richard Barr, president and CEO of MGS Machine (www.mgsmachine.com), Maple Grove, Minn., incorporated a collaborative unit from Fanuc in a palletizer he is introducing. The collaborative palletizer can handle payloads of up to 77 lbs. and operates without fencing, shrinking the footprint in half, according to Barr.
Pharmaceutical and medical device manufacturers are MGS’s primary market. The collaborative palletizer isn’t washdown-ready, but the design meets life science GMPs in terms of cleanability and the absence of niches and other harborage points, says Barr. He expects the new machine to find “nichey applications” in food & beverage facilities where very high speeds aren’t necessary and space limitations preclude use of a conventional robotic palletizer.
His palletizer can move up to six cases per minute. As with all collaborative robots, speed is a function of the types of safety devices integrated into the design. Four classifications of safety devices were created by ISO for collaborative robots: speed and separation monitoring, force and power monitoring, hand guarding and safety-rated stops.
Machine speed and payload determine the level of monitoring necessary, Barr points out, as well as the degree of human interaction. If pallet infeed and removal is done manually, a risk assessment might conclude that a safety-area scanner that slows and stops the arm is required.
The palletizer arm is covered in soft foam, more as a pinch protector than a cushion against a human collision. It’s also safety green to visually differentiate it from Fanuc’s yellow-painted robotic arms.
“Every Fanuc robot can be a collaborative robot,” emphasizes Greg Buel, product manager for collaborative robots at Rochester Hills, Mich.-based Fanuc America Corp. (www.fanucamerica.com). The difference is in the on-board safety systems, the degree of human interaction and the safety assessment of a certified integrator — and the color coding.
An app for old robots
Repurposing a robot often hinges on building or sourcing an EOA tool tailored to the new task. Product obsolescence forces the issue and puts a premium on close collaboration with system integrators and tool suppliers. A new phone app from SAS Automation (www.sas-automation.com) in Xenia, Ohio, helps jumpstart those collaborations.
A global manufacturer of EOA tools with production facilities on three continents, SAS fabricates turnkey solutions and stocks 2,000 parts for assembly by manufacturers and machine builders. When tool design requires a part that isn’t inventoried, in-house 3D printing can fill the gap.
When food handling is involved, vacuum often is the solution, according to Trent Fisher, president and owner. That was the case when tofu cakes had to be plucked from a baking pan. “It’s almost like a cheese curd,” he says of the cakes, and part of the challenge was preventing excess liquid from being sucked into the vacuum system.
The new phone app should speed delivery time for EOA tools, but it’s really the first salvo in a plan to bring smart manufacturing to robotic tooling. “In the field, people are scratching their heads over the Industrial Internet of Things and Manufacturing 4.0 and saying, ‘What does it mean to me?’” observes Fisher. His vision is to turn tool design and parts purchasing into a cloud-based activity.
Manufacturers would upload project specs and SAS engineers would “collaborate in the cloud” with them on a solution. He imagines a time when sensors built into the tools will troubleshoot problems and upload alerts that would be delivered to customers’ phones.
“Young people coming into the field grew up with a smart phone,” Fisher points out. “The time is ripe to give them a tool to take an app to the next few steps.”
Vacuum often is involved in tools Blueprint Automation (www.blueprintautomation.com) devises for contact with food. The Colonial Heights, Va., firm used vacuum to create an “upside-down tornado” in a tool to case pack cupcakes and toaster strudel at up to 100 units a minute, according to marketing manager Robbie Quinlan. Weaker than a suction cup’s force, the tool essentially creates a void, a pocket where product is suspended during transport.
Vacuum-based solutions have helped expand the universe of robotics in food handling, JLS’s Souser concedes, but the Taylor Farms system is an example of a robotic solution that wouldn’t have been possible without an advancement such as Soft Robotics’ tentacles. And vacuum isn’t a factor in machines that also perform an inspection role. A project under development involves simultaneously case packing packages of bacon and “feeling the package to see if a leak can be detected,” he says. “In some applications, if you can’t do the inspection, robotics can’t eliminate the labor.”
Simple payback calculations become irrelevant when product formulation rules out human contact. Working with OEM partner Robert Reiser & Co., JLS designed a robotic infeed for a vacuum packaging machine tailored for tortillas. Development was driven by a request from a bakery that had removed enzymes and preservatives in its clean-label tortillas and wanted to reduce the risk of cross-contamination from human handling.
“We like to think we’re helping make food safer and people’s lives better,” says Souser. When the tortilla machine was commissioned, “the workers thanked us for making their lives better” by eliminating a tedious and injury-prone job. “That’s more and more coming into play.”
Worker gratitude doesn’t turn up on a balance sheet. Neither does reduced risk of recalls. But soft benefits, expanded functionality and more affordable components add up to a more favorable ROI from robotic automation.