Risk-Based Statistical Process Control
Also known as: Risk-based SPC, RBSPC, risk-prioritized SPC, risk-informed process monitoring
Risk-based statistical process control (Risk-based SPC) is an engineering quality method that integrates formal risk analysis — typically FMEA or a risk matrix — with statistical process monitoring to focus control chart resources on the process parameters that pose the greatest risk to product quality or system safety. Rather than applying control charts uniformly across all variables, risk-based SPC directs tighter monitoring toward high-risk, high-impact process characteristics identified through structured hazard prioritization.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use risk-based SPC when a process has many measurable variables but limited monitoring bandwidth, making uniform SPC coverage impractical or cost-prohibitive. It is especially appropriate in regulated industries (pharmaceutical, medical device, aerospace, automotive) where risk-based quality frameworks are mandated or strongly encouraged by standards such as ICH Q10, ISO 9001, or IATF 16949. It is also valuable when process failures carry safety or regulatory consequences that vary markedly across parameters. Do not use risk-based SPC as a substitute for comprehensive process understanding — skipping risk assessment and applying it informally defeats the purpose. It is also unsuitable when all process parameters carry uniformly critical consequences, in which case full-coverage SPC is more appropriate.
Strengths & limitations
- Allocates finite monitoring resources to where they create the most value — reducing quality risk rather than generating charts for their own sake.
- Aligns SPC practice with regulatory risk-based quality frameworks (ICH Q10, ISO 31000, IATF 16949), easing compliance demonstration.
- Improves team focus by distinguishing critical from non-critical signals, reducing alert fatigue from low-consequence out-of-control signals.
- Creates a documented, defensible rationale for monitoring intensity decisions, which is useful during audits and regulatory inspections.
- Scales effectively in complex manufacturing environments with dozens to hundreds of potential control variables.
- Quality of the approach depends entirely on the accuracy and completeness of the upstream risk assessment — garbage-in risk scores yield misallocated monitoring.
- Requires cross-functional expertise in both SPC methodology and risk analysis, which may be scarce in smaller organizations.
- Risk scores (especially FMEA RPN) are ordinal composites that can be gamed or inconsistently applied across teams without rigorous calibration.
- Low-risk parameters that receive reduced monitoring may still drift out of control undetected if risk classifications are incorrect or process conditions change.
- Periodic review cycles may not keep pace with rapid process changes in dynamic manufacturing environments.
Frequently asked
How does risk-based SPC differ from standard SPC?
Standard SPC applies control charts to selected process variables based primarily on measurement feasibility and historical practice. Risk-based SPC adds a formal upstream step: a structured risk analysis (FMEA, risk matrix, or hazard analysis) that ranks parameters by their risk score, and then uses those rankings to determine which parameters receive intensive charting, which receive lighter monitoring, and which may be sampled periodically. The statistical tools themselves are identical; what differs is the principled, risk-driven rationale for choosing where and how intensively to apply them.
Which risk scoring system should I use — FMEA RPN or something else?
FMEA Risk Priority Number (RPN = Severity × Occurrence × Detection) is the most widely used in manufacturing, but it has known limitations: equal RPNs can arise from very different failure profiles, and the multiplicative scale is not ordinal. Alternatives include criticality analysis (Severity × Occurrence only, per MIL-STD-1629A), risk matrices (likelihood vs. consequence grids), and Bayesian belief networks for complex dependencies. Choose the method best suited to your industry standard and data availability, and document the choice in your quality plan.
How often should I update the risk assessment that drives my monitoring plan?
At minimum, review the risk assessment annually and whenever a significant process change, out-of-control event, customer complaint, or near-miss occurs. In regulated industries, change control procedures often mandate a risk re-evaluation before implementing process modifications. Treat the risk assessment as a living document linked to your control plan, not a one-time qualification exercise.
Can I use CUSUM or EWMA charts instead of Shewhart charts for high-risk parameters?
Yes, and often this is recommended. CUSUM (Cumulative Sum) and EWMA (Exponentially Weighted Moving Average) charts are more sensitive to small, sustained process shifts than Shewhart X-bar charts. For high-risk parameters where even a modest drift in the process mean could produce non-conforming product, CUSUM or EWMA charts reduce detection lag at the cost of slightly more complex setup. They are particularly appropriate when the economic consequence of a missed small shift is high.
Is risk-based SPC required by regulatory standards?
No standard mandates the exact label 'risk-based SPC,' but several frameworks require or strongly encourage the underlying logic. ICH Q10 requires pharmaceutical manufacturers to use risk management to support process monitoring intensity. IATF 16949 requires control plans that reflect risk-based thinking. ISO 9001:2015 embeds risk-based thinking throughout the quality management system. In practice, auditors in these sectors expect documented justification for monitoring decisions, which risk-based SPC provides.
Sources
- Montgomery, D. C. (2020). Introduction to Statistical Quality Control (8th ed.). Wiley. ISBN: 978-1119399308
- Statistical process control. Wikipedia. link ↗
How to cite this page
ScholarGate. (2026, June 3). Risk-Based Statistical Process Control. ScholarGate. https://scholargate.app/en/experimental-design/risk-based-statistical-process-control
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Control chartExperimental design↔ compare
- Failure Mode and Effects AnalysisExperimental design↔ compare
- Process Capability AnalysisStatistics↔ compare
- Risk-based failure mode and effects analysisExperimental design↔ compare
- Six Sigma DMAICQuality Management↔ compare
- Statistical Process ControlExperimental design↔ compare