Anchor case study · anonymisedTier-1 personal care · multi-site rollout

Tier-1 personal-care line: 960 additional bottles/hr, no new equipment

Production sits inside a contract manufacturing partner of one of the world's largest fast-moving consumer goods companies — a Top-5 global FMCG with €60B+ revenue and personal care brands sold in 190+ countries. Customer name disclosed under NDA on the call. Sector, geography, line OEMs and outcomes documented below

The setup

A global FMCG, its strategic European contract manufacturer, and the line that ships into 12 markets

The brand owner
Top-5 global FMCG
€60B+ revenue · 400+ brands · 190+ markets · personal care, home care, beauty & wellbeing
The contract manufacturer
Polish CDMO running multiple Unilever-customer facilities · 30+ packaging lines · multi-SKU personal care & cosmetics
The line
Personal care · ~150 PPM
High-mix shampoo & conditioner filling. Unilogo filler + FlexiCapper + FlexiLabeller, TransNova RUF palletizer. Multi-shift, multilingual operator team
01 /The deployment

Six things the team did

High-mix shampoo and conditioner filling, ~150 PPM, multi-shift, multilingual operator team

The line

Unilogo filler + FlexiCapper + FlexiLabeller, TransNova RUF palletizer downstream. Multi-shift, multilingual operator team

The challenge

Cap-sorter rejection rate climbing, vacuum-driven micro-stops on the delta robot, changeover variance ±18% between day and night shifts

What we deployed

Knowledge layer (auto-generated micro-lessons in 3 languages), CBM on capper torque and vacuum, Roberta pilot on weak-point monitoring

Result · throughput

960 additional bottles/hour recovered

Result · changeover

Variance compressed from ±18% to ±6%

Result · onboarding

New-operator onboarding from 14 weeks to 5

02 /The numbers

What changed in the first 6 months

+960
bottles/hour recovered
Cap-sorter throughput restored to nameplate. No new hardware
31%
line performance recovery
Versus baseline measured 6 weeks before go-live
17 min
speed loss eliminated /hr
Hidden gap between nameplate & actual run rate
14→5
weeks · operator onboarding
Time from new hire to full SKU portfolio
±18% → ±6%
changeover variance · day vs night shift
6,000+
bottles saved · per proactive maintenance fix
3 langs
operator knowledge searchable in PL, EN, UA
03 /How we got there

Six-month rollout · contract to compounding gains

No big-bang. The line stayed in production while PackGuru installed, calibrated, and learned. Phased so each module had to earn its place before the next one went in

Weeks 1–8

Knowledge layer

Captured every SOP, machine alarm code, and changeover recipe into a searchable graph. Auto-translated into PL/EN/UA. Operator queries answered in under 3 seconds at the HMI

Weeks 9–16

Condition-based monitoring

OPC-UA telemetry from the FlexiCapper torque profile and vacuum pump pressure. PackGuru flagged seal degradation 4 days before the failure that would have stopped the line

Weeks 17–24

Roberta pilot · weak-point monitoring

Roberta stationed herself at the cap-sorter — the line's documented weakest stage. Caught 11 micro-stops on day one. Live video stream to the maintenance lead's phone for tele-support

Quote

The knowledge that used to live in the heads of three senior operators is now searchable by anyone on shift, in their language

— Plant manager · attribution withheld under NDA
04 /Why this is replicable

The same loss patterns appear on every multi-SKU personal care line

Cap-sorter rejection drift, vacuum-driven micro-stops, changeover variance between shifts. We see them every time. The fixes compound — and they don't require new capital equipment

See if your line has the same pattern

Send us your sector and your top loss line. We'll come back with the closest comparable case