Examples
Examples
Try the starting example
- Batch defect counts [batch ID,pieces inspected,pieces with any defect]
- [["A",100,5],["B",80,8]]
- Batch observed defect proportions and Wilson95 bounds
- [["A",0.05,0.0215436791544,0.111750469232],["B",0.1,0.0515476155674,0.185106888061]]
- Pooled observed defect fraction
- 0.072222
Try another scenario
- Batch defect counts [batch ID,pieces inspected,pieces with any defect]
- [["A",100,10],["B",80,12]]
- Batch observed defect proportions and Wilson95 bounds
- [["A",0.1,0.0552291370607,0.174365661505],["B",0.15,0.0879404050616,0.244132285613]]
- Pooled observed defect fraction
- 0.122222
Calculation method
Calculation method
Batch observed defect proportions and Wilson95 bounds
Use unique defective-piece counts, not sum overlapping defect categories.
Pooled observed defect fraction
Total defective pieces/total inspected.
Symbols and units
- a
- Batch defect counts [batch ID,pieces inspected,pieces with any defect]
Use the units shown. A list keeps the order in which you enter its values. The symbol ^ means “raised to the power of”.
Assumptions & limits
Assumptions & limits
- Observed defective-piece rate by firing batch, with marginal binomial intervals. Per-category defect frequencies must be supplied separately; multiple defects on a piece must not be double-counted as defective pieces. Quantities are deterministic calculations from supplied observations and assumptions; no external data is fetched. Supported limits: Finite inputs only; numeric fields bounded as labelled. Lists and JSON records are limited to 256 entries unless a stricter tool bound is stated. No automatic design approval, legal conclusion, treatment recommendation or safety certification.
Are these inputs private?
These calculations run in your browser. A shared link contains the values you choose to share. Only questions submitted to the separate AI finder are sent to OpenAI.