Day in and day out, over and over, workers at a cereal distributor had to pick, twist and drop to move cereal cartons out of a bin and onto a “carton erector” conveyor. Repeatedly removing a batch of 40-50 cartons at a time from the four stacks piled in the bin made this task among the least desirable in the plant. People had a hard time with the repetitive motion, heavy loads, and awkward reach at the demanding speeds of up to 155 cartons per minute. Numerous injuries were reported and health issues were raised with this ergonomically challenging job.
>> Communicating Robots: Kuka Robotics explains its use of FDT to address complex robot bus scenarios across multiple gateways. http://bit.ly/11LbwOk
An individual in the plant who had past experience with robotics recognized that this type of operation might be a good fit for automation. JMP Engineering, a member of the Control System Integrators Association (CSIA), was called in to thoroughly investigate the process—line rates, types of cartons, accuracy, bin changeout, etc. JMP’s engineers knew that this process was a fit for automation but would require thorough testing and prototyping.
Prototype simulates gripping
The biggest challenge was properly gripping and removing the cereal cartons from the bin—at production line speed—without damaging the cartons. During the prototyping phase, JMP’s engineers tried different gripping strategies to remove bundles of cartons from the bin, testing suction cups, rubber fingers and steel plates. Through this extensive trial and error, JMP came up with a very reliable gripping mechanism that allowed them to remove 40-50 cartons at a time from the bin. JMP was confident in the solution and called the plant to tell them they had come up with a successful gripping strategy. After seeing the gripping solution, the plant’s management gave JMP the green light.
JMP went to work on implementation of the complete material handling system to remove cartons from the bin and place them on the carton erector. To do this, the gripper was attached to a Motoman HP50 material handling robot. The next challenge was to figure out how to find the four stacks of cartons in the bin—where (side to side) was each stack and how tall was it?
To solve this, JMP attached a laser photoeye to the robot, and the robot searched the bin with its laser to find the sides and tops of the stacks. This type of automation was new to the plant, so JMP worked closely with the plant’s operation and maintenance staff to make sure everyone was comfortable with the system.
Judy to the rescue
The initial startup and production period went well. The robot provided production rates well in excess of the 155 cartons per minute required with greater reliability than a person could provide. In addition, the plant was able to increase production because the robot did not take production or coffee breaks. The plant employees loved the robot so much they gave it a name, Judy, which is stamped in yellow letters up the arm of the robot. Judy continues to pick, twist and drop, and has handled millions of cereal cartons.
The client has emphasized the strength of the JMP team during this project—the thorough upfront investigation, prototyping, implementation and level of attention paid to training operations and maintenance. The robotic automation system provided all of the financial and safety benefits the client wanted.
>> Kevin Ackerman, P.Eng, is a machine vision specialist at JMP Engineering (www.jmpeng.com). He is responsible for designing and programming tough-to-automate systems with machine vision, robotics, PLCs and HMIs.
