Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
ISO 26303:2022 — Short-Term Capability Evaluation of Machining Processes on Metal-Cutting Machine Tools
A comprehensive guide to assessing machine tool process capability using statistical methods
Introduction to ISO 26303:2022
ISO 26303:2022 specifies a unified procedure for the acceptance testing of metal-cutting machine tools based on their short-term process capability. This approach, known as indirect testing, focuses on evaluating a machine tool’s ability to produce parts within specified tolerances by machining a sample batch of test pieces and applying statistical analysis.
Key Concept: Short-term capability evaluation is fundamentally different from direct testing methods (e.g., geometric or positioning accuracy tests per ISO 230). While direct testing identifies specific error sources on a machine tool, short-term capability evaluation proves that the machine can fulfill a specific production task — making it ideal for workpiece-dependent special-purpose machines and transfer lines.
The standard introduces two primary capability indices: Cs (short-term capability index) and Csk (critical short-term capability index). Cs measures the ratio of the specified tolerance to the process standard deviation, while Csk additionally accounts for the position of the process mean relative to the tolerance zone center. These indices correspond to the process performance index (Pp/Ppk) defined in ISO 3534-2 but maintain the term “short-term capability” as it has been widely used in the machine tool industry for decades.
Test Procedure and Statistical Framework
Pre-Test Agreements and Preparation
Before the acceptance test begins, the manufacturer/supplier and the user must reach agreements on several critical parameters: the workpiece features to be measured, test conditions including ambient temperature variation ((pm 3^circ)C during the test period with a maximum gradient of (pm 2^circ)C/h), and the characteristic values for evaluation. A minimum of 50 workpieces must be manufactured in series, with the total manufacturing time not exceeding 8 hours. In special circumstances, a minimum of 30 workpieces may be agreed upon.
Critical Consideration: The standard requires that the short-term capability of the measuring device itself be verified before testing. The measuring system must meet the condition (6 imes s_g leq 0.15 imes T) (where (s_g) is the measurement equipment standard deviation and (T) is the feature tolerance). This ensures that measurement uncertainty does not dominate the capability assessment.
Warm-Up and Adjustment
A proper warm-up procedure ensures the machine tool reaches thermal equilibrium before testing. If thermal distortion remains a concern, the manufacturer and user must agree on a permissible trend before running the test. The adjustment run centers the process on the target value — ideally the middle of the tolerance zone for two-sided tolerances. If the mean value shifts by even a quarter of the tolerance, the remaining 6-sigma area usable for production shrinks to only 30% of the tolerance when a Csk of 1.67 is required.
Statistical Analysis Methods
The standard provides a comprehensive statistical framework including:
| Parameter | Symbol | Formula/Definition | Application |
|---|---|---|---|
| Short-term capability index | Cs | (C_s = rac{T}{6hat{sigma}}) | Normal processes with centered distribution |
| Critical short-term capability index | Csk | (C_{sk} = minleft( rac{USL – ar{x}}{3hat{sigma}}, rac{ar{x} – LSL}{3hat{sigma}} ight)) |
Processes with shifted mean |
| Short-term range value | RV,s | (RV,s = rac{x_{max} – x_{min}}{T}) | Alternative when normal distribution cannot be assumed |
| Critical short-term range value | RV,sk | (RV,sk = maxleft( rac{USL – ar{x}}{T}, rac{ar{x} – LSL}{T} ight)) |
Range-based with location consideration |
The standard recommends threshold values for process acceptance: Cs (geq) 1.67 and Csk (geq) 1.67 for normal processes and features such as diameter or length in uncontrolled processes. For roughness values, an 80% short-term range value is typically sufficient, while in-process measurement control allows using the full tolerance.
Engineering Insights for Implementation
Managing Thermal and Tool Wear Effects
Thermal distortion is one of the most significant factors affecting short-term capability. During the warm-up phase, thermal drift of up to 40 (mu)m/h can be expected. The standard permits evaluating thermal distortion separately and agreeing on permissible limits. Tool wear must also be accounted for — if tool life substantially exceeds the manufacturing time during acceptance testing, its evaluation may be omitted. However, uncoated cutting tools in mint condition should never be used directly, as their high initial wear significantly increases cutting forces and distorts results.
Best Practice: When implementing short-term capability evaluation for multi-spindle machines or transfer lines, each spindle or clamping unit should be evaluated separately. The standard recommends using a Cs value calculated with pooled standard deviation across all units. If individual units fail, the short-term range value (RV,s) calculated from all workpieces must still be within limits to ensure all parts remain in tolerance.
Outlier Management and Distribution Considerations
The standard includes provisions for trend correction and outlier management. If the data does not follow a normal distribution — which is common in real-world machining — the standard allows using short-term range values (RV,s and RV,sk) instead of capability indices. Control charts for individuals and x̄-s control charts are recommended supplementary tools for visualizing process behavior.
Economic Implications
Choosing appropriate Cs/Csk values has significant economic consequences. While higher capability indices guarantee more reliable production, they demand substantially greater expenditure — such as direct measuring systems, probing devices, thermal compensation circuits, or switching to more expensive manufacturing methods (e.g., grinding instead of turning). The standard explicitly cautions against setting uniform boundaries for all processes without considering technical possibility and economic feasibility.
Frequently Asked Questions
What is the minimum sample size required for ISO 26303 testing?
The standard requires a minimum of 50 workpieces manufactured in series. Under special circumstances, the manufacturer and user may agree on a minimum of 30 workpieces, but statistical uncertainty increases significantly with smaller sample sizes.
How does ISO 26303 differ from direct machine tool testing per ISO 230?
ISO 230 direct testing evaluates individual machine properties like geometric accuracy or positioning error, helping identify specific error sources. ISO 26303 indirect testing evaluates the machine’s ability to produce conforming parts — it is product-focused rather than machine-focused, making it more suitable for production acceptance but less suited for diagnostic improvement.
Can ISO 26303 be applied to any type of machine tool?
The standard is primarily intended for machine tools used in large batch production with cycle times under 10 minutes. It can be applied to universal machine tools like machining centers if they meet the statistical requirements, but the standard notes that short-term capability evaluation is most appropriate for workpiece-dependent special-purpose machines and transfer line stations.
What happens if the machine fails the short-term capability test?
If Cs/Csk values or range values exceed specified tolerances, the standard requires investigating root causes — which may include outlier values, thermal instability, tool wear issues, or measurement system inadequacy. If improvements are possible, corrective actions should be implemented and the tests partially or fully repeated.
