Calculator
Example data
| Finding ID | Category | Impact | Occurrence | Detectability | Weighted Score | RPN | Severity |
|---|---|---|---|---|---|---|---|
| EX-1001 | Safety | 10 | 3 | 8 | 8.57 | 240 | Critical |
| EX-1002 | Dimensional | 7 | 6 | 5 | 6.2 | 210 | Major |
| EX-1003 | Visual | 3 | 4 | 3 | 3.35 | 36 | Minor |
Formula used
How to use
- Enter product, batch, and inspection context for traceability.
- Describe the finding and choose the best defect category.
- Score Impact, Occurrence, Detectability, Compliance, and Cost.
- Pick a scoring model, then tune weights and thresholds.
- Submit, review severity, then export CSV or PDF.
Severity consistency and inspector alignment
Consistent severity selection reduces variation between inspectors and sites. This calculator standardizes decisions by converting qualitative judgement into numeric scores. A 1–10 scale lets teams map existing defect definitions, while weights emphasize what matters most. For example, medical or aerospace programs usually raise compliance and impact weights, while high‑volume consumer lines may emphasize occurrence and detectability. Calibration workshops, using the same photos and samples, help teams align their scoring anchors.
Traceable records for audits and reporting
Structured inputs create an auditable trail for quality records. Capturing stage, batch, sample size, and a unique finding ID supports traceability during containment and disposition. When you export results, the saved inputs and the computed metric can be attached to nonconformance reports, supplier claims, or internal audits. This improves repeatability and shortens review cycles in MRB discussions. It also enables simple KPI reporting, such as Critical rate per thousand units inspected.
Choosing the best scoring model
The scoring model you choose should match how your organization manages risk. Weighted scoring is ideal when multiple dimensions must be balanced, such as regulatory exposure, scrap cost, and customer impact. RPN is helpful for FMEA‑style triage where impact, frequency, and detection are the dominant drivers. The hybrid option blends both, keeping results on a 0–10 scale for easier communication. In the hybrid method, RPN is normalized by dividing by 100 so it fits the same scale.
Setting thresholds with real inspection data
Thresholds translate the primary metric into actionable categories. Start with historical defect data, then set Minor near your median, Major near your 75th percentile, and Critical near your 90th percentile. Review monthly: if too many findings become Critical, refine weights, adjust detectability guidance, or split categories to prevent escalation fatigue. Keep written definitions for each rating point, especially for detectability, so “hard to catch” means the same everywhere. Document changes in the log.
Turning severities into decisions and actions
Use results to drive practical actions, not just labels. Critical findings should trigger immediate containment, shipment hold, and rapid root cause. Major findings typically require quarantine, rework plans, and increased sampling. Minor findings are best tracked for trend signals, and Observations can feed preventive improvements. Over time, compare severity distribution by line, supplier, and defect family. When a category escalator is enabled, treat safety and functional issues more conservatively.
FAQs
How do I choose a scoring model?
Use weighted scoring when compliance, cost, and multiple dimensions matter. Use RPN for classic triage focused on impact, occurrence, and detectability. Choose hybrid when you want RPN influence but prefer a 0–10 score for reporting.
What does Detectability represent?
Detectability reflects how difficult the defect is to catch before release. Score higher when detection requires specialized gauges, destructive tests, or rare conditions. Score lower when checks are automatic, frequent, and highly reliable.
How should we set thresholds initially?
Start from historical findings. Place Minor near the median score, Major near the 75th percentile, and Critical near the 90th percentile. Review after a few weeks and adjust to match capacity for containment and corrective action.
Why does the hybrid method divide RPN by 100?
RPN ranges up to 1000 on 1–10 inputs, while the weighted score is 0–10. Dividing RPN by 100 normalizes it to a similar scale, so combining the two metrics stays interpretable.
Do exports include all inputs and results?
Yes. CSV and PDF exports include context fields, rating values, selected model, computed metrics, severity, and recommended action. Run the calculator first, then use the download buttons to capture the most recent calculation.
Can this support audit and MRB reviews?
Yes. Consistent scoring plus saved context creates an auditable decision trail. Reviewers can see the metric used, thresholds applied, and actions recommended, reducing debate and improving repeatability across shifts and sites.