The factory is roughly 400 square metres in an industrial estate on the outskirts of Pune. It makes precision machined components for two automotive tier-one suppliers and one aerospace subcontractor. It employs thirty-one people. Two years ago it had no automation of any kind.
Today it has four collaborative robots. The owner will tell you the decision had nothing to do with strategy and everything to do with a quotation he could not refuse.
What changed on price
Cobot pricing in India has fallen by roughly 40 percent over four years, driven by three things at once: Indian assembly operations for global brands, domestic entrants competing aggressively at the low end, and a used and refurbished market that has matured into a legitimate procurement channel.
A six-kilogram payload cobot that cost the equivalent of a senior engineer’s annual salary in 2021 now costs less than a mid-level one. That crosses a threshold. It moves automation from a capital project requiring board approval to an equipment purchase a plant manager can justify against a single bottleneck.
The leapfrog
Mid-tier Indian manufacturers are not following the automation path that Western and East Asian factories took. They skipped a generation.
The traditional sequence was: dedicated automation first — hard tooling, transfer lines, fixed automation designed for high volume and one part number. Then flexible automation, once volumes justified it. Then collaborative automation, as a way of extending flexibility to lower-volume work.
- Most mid-tier Indian plants never had volumes for dedicated automation
- They cannot amortise hard tooling across the part variety their customers demand
- They are going directly to collaborative cells because it is the first automation that fits their actual production profile
That is the leapfrog. They are not behind and catching up. They skipped the stage that would have been wrong for them.
Collaborative robots priced for the workshop floor are pulling thousands of small Indian factories past a generation of automation they never had.
What the cells actually do
The Pune plant runs its cobots on three tasks, and the pattern repeats across the sector:
Machine tending. The dominant application by a wide margin. A cobot loads and unloads a CNC machining centre, which lets one operator supervise three or four machines instead of one. The payback calculation is straightforward and the risk is low.
Deburring and finishing. Consistency matters more than speed here. A human deburring a casting produces variation across a shift; a cobot produces the same edge every time. Aerospace subcontractors increasingly require it.
Inspection gauging. Loading parts into a CMM or vision station on a repeatable cycle. Unexciting, and it removes a bottleneck that used to idle the machining centres.
The constraint is skills, not capital
Every plant manager interviewed for this piece named the same problem, and it was not financing. India’s development banking and equipment leasing market handles cobot finance well — the monthly payment is comparable to one operator’s cost.
The problem is integration. A cobot arrives as a robot arm. Making it into a working cell requires fixturing, safety assessment, process programming, and maintenance capability. The supply of integrators serving mid-tier plants is thin, and the plants that succeed are usually the ones that built that capability in-house.
Two of the four cobots at the Pune plant were integrated by a maintenance technician the owner sent on a six-week programme. That technician is now the most valuable person in the building.
What it adds up to
Individual deployments are small. Four cobots in a thirty-one-person plant does not register in national statistics. Multiply across the estimated 8,000 to 12,000 mid-tier precision manufacturing units in the Pune, Chennai, and Bengaluru corridors and the aggregate is substantial.
More important than the aggregate is the capability shift. Plants that automate a bottleneck learn what automation can do. They come back for the second cell faster than the first, and the third faster than the second.
India’s mid-tier manufacturing sector is not becoming a robotics leader. It is becoming a robotics user — which, at this scale, is the more consequential thing.