Industrial robots are getting cheaper: what it means for manufacturers
The cost of industrial robots has been falling for several years. A look at which manufacturing operations this is now a practical question for.
Over the past several years, the cost of industrial robots has been declining steadily. This is not a sudden drop but a gradual trend - and by 2014 it has become substantial enough to shift the economics of automation for mid-size manufacturers.
Previously the conversation about robotisation was the domain of large automobile or electronics producers, with volumes that could justify expensive equipment. That is no longer so clear-cut.
What has changed in the economics of robotics
Several factors are working in parallel.
Prices on standard industrial manipulators have come down. This does not mean robotisation has become cheap in absolute terms - but the threshold where a reasonable payback period begins has moved lower.
A category of so-called collaborative robots has emerged - lighter, easier-to-programme devices that can work alongside people without heavy safety enclosures. This opens possibilities for production environments where replacing a person entirely is not the goal, but automating specific operations is realistic. The earlier question of how to count the value of warehouse automation beyond headcount applies equally to the factory floor.
The cost of integration and programming has not fallen as sharply as the hardware. This is an important nuance: the equipment itself is more accessible, but the work of implementing it still requires skilled people.
Which tasks this is relevant for right now
Not all manufacturing tasks are equally suited to robotisation. The best candidates share a few characteristics.
The operation repeats at high frequency with little variation. Welding standard seams, loading and unloading machines, assembling identical units - good candidates. Tasks requiring judgement or adaptation to non-standard situations - not good candidates.
There are physical limitations for humans: hazardous conditions, high temperatures, repetitive loads that lead to injury. Here robotisation addresses both an economic and a human problem. The same production environments where these constraints are sharpest are also where machine vision for quality control starts to earn its place alongside the robot arm.
Volume is sufficient for payback. Even at lower prices the capital investment remains significant. If an operation runs one hour a day, the economics do not work.
What stands between intent and result
A common misconception is to think about a robot as a one-to-one replacement for one person. In practice, robotisation changes the organisation of work more broadly.
You need a specialist who maintains the equipment and can reprogram it when the task changes. You need stability in incoming materials - a robot will not tolerate the same variation that an experienced worker handles by feel. You need processes for handling abnormal situations that the robot does not recognise.
Manufacturers who have seen good results from robotisation typically started with one specific operation, went through the full cycle - from technical specification to stable operation - and only then expanded the approach to other areas.
Questions to assess readiness
If you are considering robotising a particular operation:
- How standardised is the operation - is there a clear specification of exactly what the robot should do?
- What is the volume of this operation in hours and units per year?
- Who on the team will be responsible for maintenance and reprogramming?
- How consistent is the incoming material or component?
- How will the organisation of work at this station change - what do people do while the robot works?
Without answers to these questions, any economic justification will be imprecise and the risk of a disappointing pilot is high.