MFU vs. PFU: Why Machine Capability Alone Says Nothing About Your Screwdriving Process

Imagine this: your screwdriver passes the machine capability study with an excellent Cmk value of 1.67 - exactly the threshold the automotive industry requires for safety-critical joints. The team is satisfied, the documentation complete, the process released.

And then: customer complaint. Faulty screw joints from series production.

Contradictory? Not at all. This scenario occurs regularly in practice - and is almost always due to the same flawed assumption: the MFU (machine capability study) is treated as proof of process quality. It is not. And this misunderstanding can become very expensive.

This article explains what MFU and PFU actually measure, why the difference is critical in mass production, and how to apply both capability studies correctly for reliable torque control, repeatability, and assembly quality assurance.

What the MFU measures - and what it deliberately ignores

The machine capability study (MFU) according to VDI/VDE 2645 Part 2^[1] has a clearly defined task: it determines the stability and repeatability of the process influence factor "machine."

That sounds precise - and it is. But that precision comes at a price: the MFU is a short-term study carried out under deliberately idealised conditions. Disturbance variables are systematically eliminated:

• No operator change -> eliminate human influence
• No material change -> exclude batch variation
• No work interruption -> avoid shift and break effects
• Stable environmental conditions -> prevent temperature and humidity fluctuations
• No tool change -> eliminate setup influences

The result is the Cmk value - a meaningful characteristic number for machine scatter under ideal conditions. In the automotive industry, it is standard practice to require a Cmk ≥ 1.67, which corresponds to process scatter where statistically only 0.57 ppm (parts per million) of parts lie outside the tolerance.

The MFU therefore answers a very specific question that is central to any screwdriving process: is the tool fundamentally precise enough? It says nothing about what really happens at the assembly line.

What the PFU measures - the reality of series production

The process capability study (PFU) according to VDI/VDE 2645 Part 3^[2] asks a fundamentally different question: does the entire screwdriving process, under real series conditions, consistently stay within the required tolerances?

Unlike the MFU, the process capability study also takes into account the influence categories human, material, method and environment in addition to the machine influence. This is the well-known 5M model of industrial production control - and in torque-controlled screwdriving processes all five dimensions act simultaneously:

The goal of the process capability study for screw assemblies is to assess and document the quality capability of a screwdriving process under series production conditions. The core instrument for this kind of assembly testing is the residual torque measurement: after tightening, a calibrated inspection tool is used to determine the torque required to turn the screw a small additional angle. This residual torque provides information about the actually installed preload force - and thus about the real joint quality and screw assembly quality.

The result of the PFU is the Cpk value - the critical process capability index, which reflects both the position of the mean relative to the tolerance limits and the overall scatter of the process. In other words, it captures your real assembly process stability.

The crucial comparison: MFU and PFU at a glance

Practical example: Cmk 1.67 in the lab - Cpk 0.95 in production

Let us look at a realistic scenario from automotive assembly quality assurance:

An electric screwdriver is approved for a Class A joint (chassis component). The MFU under laboratory conditions yields a Cmk of 1.67 - the minimum requirement is met. The release is granted.

Three months later in ongoing production: the PFU shows a Cpk of 0.95. The process is not capable. What happened?

The analysis reveals several overlapping influence factors affecting torque analysis and process capability:

• Variations in lubricant: the supplier has changed the lubricant batch. The coefficient of friction under the screw head has shifted - the installed torque systematically deviates from the target.
• Seasonal temperature fluctuations: in winter, temperatures in the assembly hall drop below 15 °C. The screwdriver behaves differently with cold components than in the temperature-controlled lab.
• Operator rotation: different workers apply different reaction forces, which leads to measurable scatter with hand-guided tools.

None of these factors were visible during the MFU - because they were all deliberately excluded. The tool was capable. The process was not.

What Cmk and Cpk really tell you

The relationship is clear: the minimum value for machine capability (Cmk) must be higher than the threshold for process capability (Cpk), because beyond the MFU time frame additional disturbance variables act on the process and increase process scatter.

This means: a Cmk of exactly 1.67 leaves virtually no margin for the unavoidable production influences in real-world production monitoring. Achieving a robust Cpk in series production requires a significantly higher Cmk.

Try it yourself: the MFU-PFU influence simulator

How strongly do the 5M factors change your Cpk in practice? Set the disturbance variables yourself and see how the process capability value reacts in your screwdriving process:

When do you need what? The right use of MFU and PFU

The basic rule is straightforward:

• MFU = Once, during tool procurement, after every repair and after major changeovers
• PFU = Regularly in ongoing production, after process changes and following any abnormalities

A successful MFU is a prerequisite for the PFU - but not a substitute. An MFU without a subsequent PFU is like a vehicle that passes inspection on the test stand but is never driven in real traffic.

What the standards specifically require: VDI/VDE 2862 and VDI/VDE 2645-3 working together

The two sets of guidelines interact directly and form the framework for norm-compliant capability study and measurement data analysis:

VDI/VDE 2862^[3] classifies screw joints into three risk classes:

• Class A - safety-critical: risk to life and limb in the event of failure (e.g. wheel suspension, braking system)
• Class B - function-critical: product failure in the event of malfunction ("breakdown")
• Class C - non-critical: no safety- or function-relevant consequences

The classification determines the minimum requirements for tools, monitoring and documentation. VDI/VDE 2862 has applied to the automotive industry since 1999 and, since 2015, to all plant, machinery and equipment manufacturers.

VDI/VDE 2645 Part 3 (PFU) provides the procedure for demonstrating the process quality required by VDI/VDE 2862. The higher the risk class, the tighter the tolerance limits - and the more demanding the required Cpk value for process capability.

The PFU provides guidance for assessing and continuously improving the screwdriving process, including identifying systematic influences, evaluating process improvement measures and defining intervention limits for quality control charts.

Anyone who has not yet fully understood VDI/VDE 2862 and its screw joint classes A, B and C should start there - because the classification is the starting point for all capability requirements in screw assembly quality.

The right measuring tool for MFU and PFU in one

A PFU is only as robust as the measuring system used. Clear requirements apply to both types of torque control study:

• DAkkS-accredited calibration with complete metrological traceability
• Residual torque measurement without operator influence for reproducible PFU results
• Automatic Cmk/Cpk calculation with statistical evaluation and archiving
• Standards-compliant documentation of all torque curves and measurement data for audits

The GWK QUANTEC MCS® was developed precisely for this use case: as an electronic torque and angle analysis tool with reference-free angle measurement and ±1 % measurement accuracy between 10 and 100 % of the nominal range, it covers both capability study types. The integrated residual torque function provides the PFU data basis directly on the component - without returning it to the lab.

The analysis software QuanLabPro automatically evaluates Cmk and Cpk values, provides control charts and histograms, and archives all measurement data in a tamper-proof manner. This enables MFU and PFU to be implemented from a single source - with one calibrated tool, ideal for efficient torque analysis, assembly testing and ongoing production monitoring.

If you wish to avoid investing in your own measurement equipment or need an analysis tool for a time-limited PFU campaign, you can also use the Quantec MCS® flexibly via the GWK ToolRent® rental system - including verified calibration before delivery.

Conclusion: MFU is the starting point - PFU is the proof

The machine capability study is indispensable - but only the first step. It proves that your tool works precisely under ideal conditions. Nothing more.

Only the process capability study shows what really happens in production: with changing operators, fluctuating material batches, seasonal temperatures and the hundred small disturbance variables that shape everyday assembly. The Cpk value is not a bureaucratic formality - it is your earliest warning signal, before a quality problem leaves the assembly line.

Practical takeaways:

• MFU with Cmk ≥ 1.67 is mandatory for procurement and after repair - but not a free pass for series production
• PFU according to VDI/VDE 2645-3 is the standards-compliant proof of process quality in series manufacturing
• Cpk is structurally lower than Cmk - plan for this gap right from the start
• Both studies must be carried out independently and documented separately
• Every relevant process change (material, lubricant, operator, environment) requires a new PFU

If you want to carry out a PFU step by step, our guide provides a detailed roadmap to your first standards-compliant process capability study - from sampling plan through to Cpk calculation and complete measurement data analysis.