DEPLOYMENT STORY · AUTOMOTIVE9 MIN READPUBLISHED 4 SEP 2026

How Nordwerk Automotive scaled from pilot to 120 Forma robots in 16 weeks.

A door-module line in Zwickau, two shifts, no re-layout. What worked, what didn’t, and the three changes that made scaling possible.

AUTONOMOUS · 1× · UNEDITED PHOTO · Forma units at the Zwickau trim line, week 16 · honest light, wide
KITTING COST
€4.10€2.54
−38 % per kit
KIT ACCURACY
96.4 %99.6 %
Mis-picks per 1 000: 36 → 4
LINE STOPS / WEEK
71
Caused by missing parts
UPTIME
—99.2 %
90-day rolling, Fleet OS

Baseline: 12 weeks before kickoff · after: weeks 13–16 · customer-verified · illustrative

WEEK 0
0Forma units live
Pilot contract signed
WEEK 0 · KICKOFF

Success criteria before robots

Nordwerk’s plant manager asked for one thing before signing: a written definition of success. We agreed on three numbers: 95 % kitting accuracy, 20 % cost reduction per kit and zero lost-time incidents over 6 weeks.

“I have seen too many pilots that succeed because nobody defined failure.”
Katrin Albers, Plant Manager, Nordwerk Zwickau
WEEK 2 · PILOT LIVE

Two robots, one aisle, real parts

Two Forma units started moving door modules between the rack and the line-side carts on the Monday of week two. Operators set the pace; the robots followed the existing takt without changes to the layout.

WEEK 6 · CRITERIA MET

All three numbers, with margin

Accuracy reached 99.1 %, cost per kit fell 31 %, and there were no incidents. The fleet grew to eight units while we fixed the gripper-finger wear issue on sharp trim clips.

“The data came from our MES, not from Kairon slides. That made the decision easy.”
Tobias Reuter, Head of Operations Excellence
WEEK 10 · SCALE

From one aisle to three halls

With Fleet OS connected to Nordwerk’s MES, each new cell took two days to commission. The limit was never the robots; it was how fast charging stations could be installed.

WEEK 16 · STEADY STATE

120 robots, two shifts, 99.2 % uptime

By week sixteen, 120 Forma units covered kitting, line-side delivery and empty-rack return across both shifts. Six operators moved to fleet supervision and quality roles; nobody was laid off.

“Our people now run the robots. That was the part I was most worried about, and it went best.”
Katrin Albers
BEFORE · manual kitting, 6 operators per shift
AFTER · 8 Forma + 2 operators
PHOTO PAIR · same camera position, weeks 0 and 16

What made it scale

  1. 01Define failure first
    Written success criteria turned the pilot into a decision, not a demo.
  2. 02Connect to the MES in week one
    Real-time data from Nordwerk’s own system ended every debate.
  3. 03Plan infrastructure, not robots
    Charging stations and Wi-Fi coverage set the pace of scale.

What didn’t work: the first gripper finger design wore out in 9 days on sharp trim clips. The fix took two weeks and is now standard.

Your line, week one.

Start with a 6-week paid pilot. Same team, same success-criteria template Nordwerk used.

Book a pilot → Download the success-criteria template
CONVERSION · STORY
Proof format: case metrics with baselines + honest setback. Secondary is a gated resource (micro-conversion).