Pick and place: a guide to choosing and automating
Robot, vision, gripper, feeding and cycle: the decisions that define whether a pick and place cell works in production.
Pick and place looks simple: pick a part and place it somewhere else. In production, the difficulty is part variation, how it arrives, orientation, accuracy and cycle time. This guide organizes the technical and commercial decisions before requesting a quote.
Which tasks count as pick and place
Picking, orienting, placing, sorting, inspecting and assembling. Each task has different requirements: moving a box from a conveyor to a pallet is not the same as picking a randomly placed part from a bin, orienting it and assembling it with tolerance. Defining the task is the first input for choosing robot and tooling.
SCARA, delta, 6-axis or cobot
SCARA is fast in the plane and compact; delta wins on very short cycles with light parts; the 6-axis robot orients the tool in any direction and covers large volumes; the cobot brings flexibility and proximity, with payload and speed limits. The comparison uses the part, travel, orientation and cycle, not a brand preference.
How the part arrives
Trays, conveyors, feeders, boxes or random bins. Feeding defines whether vision is needed, whether a fixed position works or parts must be separated. It also defines dead time: the fastest robot loses if it waits for the part.
Vision: when it is needed
2D vision is enough for flat parts in known positions; 3D and bin picking locate stacked or randomly placed parts. Robot-camera calibration, lighting and contrast are part of the project. Vision does not replace good part presentation: it complements it.
Grippers and end effectors
Vacuum, parallel gripper, suction cups, soft gripping or dedicated tools. The choice depends on material, surface, weight, fragility and placement accuracy. Format changes and end-effector maintenance must also be planned.
Conveyor tracking, PLC and safety
If the part moves, the robot must follow it with an encoder and synchronization logic. The cell integrates with PLC, sensors and safety: fences, light curtains, scanners or a cobot with risk assessment. ISO 10218 and ISO/TS 15066 are safety references, not an automatic guarantee.
Cycle and payback with assumptions
Cycle time is calculated from pick, transfer, placement, waiting and release. Payback depends on labor, output, rework and uptime. Any published number must explain its assumptions and separate external benchmarks from first-party data.
What we need to assess your part
Photos or video of the part and how it arrives, weight, dimensions, material, orientation, placement tolerance, required rate and environment. With that data, robot, vision, gripper and feeding can be proposed, and a trial can be defined.
Common mistakes
Choosing the robot before understanding feeding, underestimating part variation, failing to plan format changes, forgetting safety or comparing quotes with different scopes. Buying vision for a problem that a better-designed tray would solve is also a mistake.
Frequently asked questions
What is pick and place?
It is the operation of picking a part, orienting it and placing it somewhere else. It can include sorting, inspection or assembly, and each task changes robot and tooling requirements.
When is pick and place not worth automating?
When the part is one-off, position changes without rules, volume is very low or feeding cannot be stabilized. In those cases, streamline the process before investing.
How does part feeding affect the cell?
It defines whether vision is needed, whether a fixed position works and how much dead time the cycle tolerates. The fastest robot loses if it waits for the part.
What is the difference between SCARA, delta and 6-axis?
SCARA is fast in the plane and compact; delta wins on very short cycles with light parts; 6-axis orients the tool in any direction and covers large volumes.
How is a cell validated before buying?
With a trial using the real part and feeding, measuring pick, transfer, placement and release. The trial also confirms tooling, safety and acceptance criteria.
What are the most common mistakes?
Choosing the robot before understanding feeding, underestimating part variation, failing to plan format changes or comparing quotes with different scopes.
