What is MCRL (Manufacturing Capability Readiness Level)?

A nine-point scale for how far a manufacturing capability has actually been taken toward making the thing repeatably, at rate, at cost and at yield. Created at Rolls-Royce to sit beside technology readiness, because knowing that something works tells you almost nothing about whether you can make it.

Where it comes from

MCRL is not ours. It was developed by Michael Ward at Rolls-Royce, published in 2012 with Halliday and Foden in the Proceedings of the Institution of Mechanical Engineers, Part B, and identified as good practice by the UK Aerospace Design and Manufacturing National Technical Committee. Rolls-Royce set the scale out to a House of Commons select committee in the same period.

The reasoning behind it was blunt. Technology readiness levels do not survive contact with a factory: they are hard to interpret precisely in a manufacturing setting, and the technology itself is only one of several things that decide whether an implementation works.

It is a British industrial framework, roughly fifteen years old, barely used outside aerospace. That is the odd part. The measure exists, it is sound, and almost nobody scores against it.

The nine levels, in three bands

• MCRL 1–4 — conception and assessment. The manufacturing route is proposed and tested at the scale of a laboratory or a single cell.
• MCRL 5–6 — pre-production. Full-scale equipment and the real process, run before the product has launched and before the factory has been paid for. This is the expensive band.
• MCRL 7–9 — implementation. The process goes onto the shop floor and volume production is confirmed with assured quality.

Rolls-Royce named the problem inside their own scale: there is a valley of death around MCRL 4 to 6, where the spend is heaviest and there is still no certainty that the product will launch or that the process will work. That is the same gap, arrived at from the engineering side, that we have argued is an accountability hole rather than a funding one — see Nobody owns the valley.

MCRL is not MRL

Two acronyms one letter apart, measuring different things for different readers.

• MRL — Manufacturing Readiness Level. Ten levels. Developed by a US Department of Defense working group, adopted in 2005 and mandatory on major acquisition programmes from 2011. It asks whether a programme is ready to be manufactured.
• MCRL — Manufacturing Capability Readiness Level. Nine levels. Rolls-Royce. It asks whether a capability is ready to make the thing.

The count matters. Nine levels map one-to-one onto the nine of TRL, which is what lets you set the two ladders side by side and read the distance between them. That distance is the Coupling Gap.

What a single MCRL number aggregates

This is where we extend it. A readiness level is a summary, and a summary is only useful if you know what it summarises. Each of the following has been separately named and separately scored by somebody, usually a different function, usually on a different form.

• Materials readiness. Is the material specified, sourced and characterised — and does batch-to-batch variation sit inside the window the process assumes?
• Process readiness. Is the process window characterised and capable, or is it a set of parameters that worked on a good day?
• Equipment and tooling readiness. Does the plant exist, is it qualified and calibrated, and can it be maintained by the people who are actually on site?
• Supply chain readiness. Have named suppliers quoted the critical steps, having made this class of part before, with a second source wherever one failure stops the line?
• Workforce readiness. Has anyone on the payroll taken this class of process from concept to rate before?
• Culture readiness. Will the organisation run the process as designed at three in the morning in week nine, or work around it and not write that down?
• Measurement readiness. From the data actually collected, can anyone tell whether the process is in control?
• Cost readiness. Is the cost per good unit at the target rate known, as against the cost of the demonstrator?

The rule we apply is that the score is the lowest of them, not the average. A capability with materials at 8 and culture at 3 is at 3. Averaging is how a programme reaches the pre-production band believing it is ready, and finds out there which one it was.

What a level is evidenced by

Across all eight, the test is the same: evidence rather than intent.

• Process evidence, not demonstration. A part made once proves the design. A qualification report proves the process.
• Yield and rework as they are on the floor, not as they appear on the dashboard.
• Cost per good unit at the target rate, not the demo unit cost.
• Supplier qualification for the critical steps, with named suppliers who have quoted.
• People: whether anyone on the payroll has taken this class of process from concept to rate before.

Why the manufacturing engineer owns it

Every one of those rows belongs to somebody in a large organisation. Materials to materials engineering, supply to procurement, people to HR, culture to nobody in particular. The manufacturing engineer is the only role obliged to carry all of them at once, because the process does not care which department the weakness came from.

That is also why this practice is named the way it is. Kaipability is the ability to change what you are capable of. A readiness level tells you what you are capable of today. The work is moving it.

Why it is rarely scored

Because the expertise to score it does not sit in venture capital, growth equity or corporate development. It sits in manufacturing engineering, and it has sat there quietly for decades while the language of readiness was borrowed by everyone else.

The result is a systematic blind spot. Capital flows toward technology maturity because technology maturity is measurable, and production readiness is not. Or rather, it is measurable — just not by the people doing the measuring.

This is the measure Deployment Readiness reports against, and the one an operator-grade read is performed to establish.