Operations & Supply Chain · Grow

Yield loop.

Reads production data against spec targets and proposes the setting changes that hold quality while costing less.

Operational lens+3.4 pts gross margin modeled targets · not client results
.crft/Yield loop

gross margin

to the Scan's evidence — before any scope

4–10 wks

to a system in production, not a pilot

modeled targets · not client results · timings are CRFT's standard engagement

The work, today.

Targets were set once, by someone who has usually moved on, with a safety margin nobody has revisited. The line runs to them because that is what the sheet says, and the cost of that margin is invisible because it has always been there.

Where the Scan startsWith production data, current spec targets and who sets them today.

What the system is.

Four lines, the same four on every system we build. Where they fall is what makes this one different from the last one.

Line and batch data at whatever granularity you capture it
Current spec targets and their tolerances
Scrap, rework and quality-hold rates
Input cost and its variability
01

What it reads

The signals it works from. Typically connects to MES or historian, Quality system, ERP and Cost data.

01Where the real quality cliff is, as opposed to the documented margin
02Which settings are over-specified for the outcome they protect
03What a change is worth, and what it risks
04Which changes are safe to trial and which are not
02

What it decides

The calls it makes on its own, unattended, every time it runs.

A ranked list of proposed changes with expected value and risk
A trial protocol for each, sized to prove or disprove it
Results back against the prediction, so the model is held to account
03

What it writes back

Into the tools the work already lives in.

04

Where the human stays

Process engineering approves every change and quality holds a veto. This is the case where the human line is least negotiable — the system proposes and predicts, it never touches a setting.

How we set it up.

The same four legs as every CRFT engagement — written for this system rather than in general.

  1. 01
    Scan48 hrs · evidence before scope

    We test whether your data can support the question at all. Sampling rate and sensor coverage decide this, and the honest answer is sometimes that the instrumentation comes first.

  2. 02
    Scope1 week · the report is the scope

    One line, one product family, one set of targets. Yield work generalizes badly across lines and should not be scoped as though it does.

  3. 03
    Build4–10 wks · production, not a pilot

    Prediction is validated against held-back history before a single change is proposed. Nothing reaches the floor on the strength of a backtest alone.

  4. 04
    Compoundongoing · it reports, then improves

    Every trial, successful or not, is a labeled experiment. The model gets specific to your line, which is exactly what a generic optimizer cannot be.

What we hold it to

Margin points against the pre-change baseline, with quality holds flat or better — both conditions, or it has not worked.

Get CRFT

Build the intelligence your company compounds on.

Five inputs, a published price, a report in 48 hours. Decide with numbers — then own what gets built.

Or just send us a note
What kind of system? Pick any that apply