A pick and place robot does exactly what its name says. It takes an object from one place and puts it in another, over and over, faster than a person and without getting bored. It is the simplest robotic application there is, which is why it is usually the first one a plant buys and the one where the cost is most often underestimated for reasons that have nothing to do with the robot.
The four numbers that specify the arm
Payload is the weight of the object plus the gripper that holds it, not the object alone, and that distinction catches people out. Reach is the distance from the robot base to the furthest point it has to touch, measured in the real layout rather than on the drawing. Cycle rate is picks per minute at the required accuracy. Repeatability is how closely the arm returns to the same point. Get those four right and the choice of machine narrows quickly.
Kinematics change the answer
A delta robot suspended above a conveyor is extremely fast over a shallow working area and suits light objects at high rate. A SCARA arm is quick and rigid in a plane and suits assembly and placement into a fixture. A six-axis arm is slower and can approach from any angle, which is what heavier or awkwardly oriented parts need. Choosing the wrong family is the most expensive mistake available, because it is not fixable with better programming.
Presentation is the real project
A robot picks reliably from a known position. Objects arriving in a heap need vision, and vision needs contrast, lighting and a shape the system can resolve, none of which is free. Objects arriving in a nest, a tray or on a conveyor with a fence and a stop can often be picked with no vision at all. It is almost always cheaper to fix how the parts arrive than to buy a system clever enough to cope with chaos.
Grippers and the parts that fight them
Vacuum is fast, simple and cheap on flat, clean, non-porous surfaces. Mechanical fingers handle irregular or porous items and take longer to close. Soft and magnetic tooling exist for the cases that defeat both. Objects that are wet, dusty, flexible, sticky or nested inside each other are where a straightforward application becomes a development project, so say so early if any of that describes yours.
Where the payback actually comes from
Straight labour substitution is the weakest argument and often the one presented. The stronger cases are consistency where a misplacement causes downstream scrap, uptime through breaks and shift changes, taking a person off a repetitive lifting task, and freeing a skilled operator to do work the robot cannot. Price the cell against those together rather than against one wage.
Questions people ask about pick and place robot
What is a pick and place robot?
It is a robot arm, often a delta or SCARA type, that repeatedly moves objects from a supply position to a destination position, typically between conveyors, trays or fixtures.
How fast can one run?
It depends far more on the object and the distance than on the arm. Short moves with light objects and a vacuum gripper are quick; long moves, heavy objects or a gripper that has to close mechanically are slower.
Do I need machine vision?
Only if the objects arrive in an unknown position or orientation. If they can be presented in a fixture, a nest or against a stop, a cell without vision is cheaper, faster and considerably easier to keep running.
What is the difference between payload and part weight?
Payload is what the robot carries at the wrist, which is the part plus the gripper plus anything else mounted there. A heavy gripper can put a light part outside the rating of an arm that looked adequate.
Can a pick and place cell handle more than one product?
Usually yes, if the products are similar enough for one gripper and one presentation method. Changeover between genuinely different products means tooling changes, and those need designing in from the start.