Heavy Palletizing: The Layout Starts at the Carton
Direct answer: A palletizing cell is planned from the carton, not from the robot. Carton size, weight and which faces can be gripped decide the gripper; the gripper’s own mass and overhang come off the payload rating; the pattern sets the pick and place points, and reach is consumed above the finished stack rather than on the picking side. The cycle is limited by the infeed rate, so a faster arm buys nothing once cartons arrive more slowly than one cycle.
Who this is for: Written for packaging and logistics engineers specifying an end-of-line palletizing cell.
Scope: This EVST guide covers the order a palletizing cell is planned in: carton first, then gripper, then payload budget, then pattern and posture, then cycle against the infeed. It does not quote cycle times, payload margins, stack heights or stability results; those come from calculation and trial running with your own product.
Start at the carton
The first inputs to a palletizing layout are properties of the product, not of the robot: outer dimensions, unit weight, which faces are available to grip, and whether the packaging will take clamping at all.
In the reference footage the gripper takes a full carton from an infeed conveyor, turns and places it on a pallet, and the stack grows course by course through the sequence. The operation looks simple; the gripper form behind it is not a free choice.
Clamp, vacuum, fork or a combination follows from the packaging. A carton with a weak top flap rules out vacuum on that face; a bag or a drum rules out clamping geometry designed for a box. Change the product and the gripper changes, and every downstream number changes with it.
The order EVST works in is the same each time: carton, then gripper, then payload budget, then pattern and posture, and the cycle against the infeed last.
The payload budget for heavy-payload robotic palletizing
A robot’s rating is what the flange can carry in total. The gripper is part of that total, so what remains for the product is the rating minus the tool.
Palletizing tools are rarely light. A full-layer frame with a backing plate, pneumatics, sensing and cushioning adds up quickly, and a larger head always leaves less for the load it was made larger to handle.
Mass is only half of it. The further the tool’s centre of gravity sits from the flange, the more it loads the wrist at the same weight, and inertia rises with it. A check that considers weight but not overhang passes on paper and produces alarms at working speed.
According to ISO 9283:1998, pose repeatability is measured under stated load and speed conditions, which is why a repeatability figure quoted without the tooling and product mass it assumes does not describe the cell you are buying.
The pattern sets the posture, and the top of the stack sets the reach
Only once the stack pattern is fixed can the pick and place points be defined. In the footage the pattern is visible in how the courses interlock and in how the placement points move as the stack rises.
The bottom course and the top course are different postures, not the same posture at different heights. Low placements require the arm to reach down and in; high ones require the wrist to clear the courses already placed.
That is where reach is consumed. Layouts that look comfortable measured at the pick point run out at full stack height, and by the time that is discovered the cell is built. Check the envelope at the last carton of the last course first.
| Input | What it decides | What to verify first |
|---|---|---|
| Carton size, weight, grip faces | Gripper form | Whether the packaging takes clamping or vacuum at all |
| Gripper mass and overhang | Usable payload and wrist load | Centre of gravity, not only total mass |
| Stack pattern and height | Pick and place points and posture | Envelope at the top course, not at the pick point |
| Infeed rate and pallet change | Achievable cycle | Whether the robot waits on the line or the line waits on the robot |
The cycle belongs to the infeed
In the footage one carton arrives and the robot performs one cycle. Stated that plainly it sounds trivial, but it is the ceiling on the cell’s output.
The arm’s top speed determines how long its own cycle takes. If cartons arrive at a longer interval than that, the arm waits, and a faster model changes nothing about the line rate.
So the cycle calculation starts upstream: arrival rate, whether a full layer is accumulated before transfer, how long a pallet change takes, and how finished pallets are removed. Those together are the real output, and any one of them can be the constraint.
Product stability matters here too. According to ISO 12048:1994, the compression resistance of shipping containers is determined by a stated test method, so how high a carton can be stacked is a packaging question answered by testing rather than by the robot specification.
What belongs to calculation, and safety
Achievable cycle time, remaining payload margin and stack stability in transport come from calculation and trial running. None of them can be read from footage of a cell operating, and none are quoted here.
According to ISO 10218-2:2025, the safeguarded space of an integrated cell follows from the movement of the whole system, which for palletizing includes the load in the gripper and its swept volume, plus pallet and product entry and exit.
According to ISO 12100:2010, dropped loads, pallet handling, access for jam clearing and the interface with the conveyor belong to the same risk assessment as the robot motion.
Every EVST application review opens from the same project inputs – carton data, pattern and infeed rate – so selection rests on evidence and acceptance is agreed against your own product rather than a datasheet.
Frequently asked questions
Why not start from the payload and reach columns?
Because both are outputs. The carton and the pattern decide the gripper, the gripper decides how much of the rating is left, and the top course decides how much reach is actually needed. Starting from the datasheet usually produces a model that fits the pick point and not the finished stack.
How much payload does the gripper itself take?
Enough to matter, and it depends on the form. A full-layer frame with backing plate, pneumatics and cushioning is substantially heavier than a single-carton clamp. Its centre of gravity has to be considered as well, because overhang loads the wrist beyond what mass alone suggests.
Will a faster robot increase output?
Only if the robot is the constraint. If cartons arrive more slowly than one robot cycle, the arm is already waiting and a faster one changes nothing. Work out the infeed rate, layer accumulation and pallet change time before comparing models.
What should we send to get a useful answer?
Carton dimensions and weight, which faces can be gripped, the stack pattern and height, the infeed rate, and how finished pallets leave the cell.
Project inputs for an application review
To have this checked against your own product rather than a generic cell, send:
- carton dimensions, unit weight and available grip faces
- stack pattern and finished stack height
- infeed rate and whether a layer is accumulated first
- pallet change and removal method
Send the carton size, weight and pattern and the gripper form and model can be worked through against them. Related reading: industrial robot product range, automated factory integration.