ExampleConstruction: AI-generatedVerification: AI-generatedPipeline-generatedprecheck passjudge pass (gpt-6-luna)

How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Day-specific calibration slope drift and check standards

Example

Suppose the three stable reference standards have assigned quantities $q \in {0, 5, 10}$ . Remeasure these same standards on two days, obtaining the stipulated exact response table

Reference quantity $q$	Day 1 response	Day 2 response
$0$	$1$	$1$
$5$	$11$	$12$
$10$	$21$	$23$

Take $[0, 10]$ as the declared calibration range. The standards are check standards here: fixed reference artifacts measured again over time to reveal changes in the measurement process. The numbers are a deterministic illustration, not observed data. Fit one equally weighted straight line with an intercept to all six response pairs and compare it with the separate day-specific calibration curves.

Facts & Assumptions

Given: The assigned reference values and stipulated responses in the Example, with the pooled line defined to minimize the sum of squared vertical residuals over all six pairs.

[F1]

Measurands, calibration, and measurement uncertainty: a calibration curve is a specified relation between a reference quantity and instrument response used to convert a later response into a measurand estimate.

[F2]

Batch and measurement-process variation: day is a batch/process condition, and calibration curves are pooled only under an explicit shared curve or equivalence model.

[A1]

The three reference standards remain stable with assigned values $0, 5, 10$ on both days, and the response table is stipulated exactly as shown.

[A2]

The pooled line is $a + b q$ with $a, b \in R$ , fitted by minimizing $\sum_{d = 1}^{2} \sum_{q \in {0, 5, 10}} (r_{d q} - a - b q)^{2}$ with equal weight for each of the six readings.

Verification

1.1givenA1algebra

The day-specific lines are $r_{1} (q) = 1 + 2 q$ and $r_{2} (q) = 1 + 2.2 q$ : each intercept is the response at $q = 0$ , and the slopes are $(21 - 1) / 10 = 2$ and $(23 - 1) / 10 = 2.2$ . In particular the known standard at $q = 5$ has responses $11$ and $12$ .

1.2givenA1algebra

The six reference values have mean $\overset{ˉ}{q} = 5$ and the six responses have mean $\overset{ˉ}{r} = 69 / 6 = 11.5$ . Their centered sums are $S_{q q} = \sum (q - \overset{ˉ}{q})^{2} = 100$ and $S_{q r} = \sum (q - \overset{ˉ}{q}) (r - \overset{ˉ}{r}) = 210$ : the two $q = 0$ readings contribute $105$ , the $q = 5$ readings contribute $0$ , and the two $q = 10$ readings contribute $105$ .

2.1givenA2step 1.2algebra

Write each residual as $(r - \overset{ˉ}{r}) - b (q - \overset{ˉ}{q}) + (\overset{ˉ}{r} - a - b \overset{ˉ}{q})$ . The first two terms sum to zero over the six pairs, so expanding the sum of squared residuals gives $\sum [(r - \overset{ˉ}{r}) - b (q - \overset{ˉ}{q})]^{2} + 6 (\overset{ˉ}{r} - a - b \overset{ˉ}{q})^{2}$ . For fixed $b$ this is minimized at $a = \overset{ˉ}{r} - b \overset{ˉ}{q}$ ; the remaining slope-dependent terms are $S_{q q} b^{2} - 2 S_{q r} b$ , whose completed square is $S_{q q} (b - S_{q r} / S_{q q})^{2}$ plus a constant. Since $S_{q q} = 100 > 0$ , the unique minimizing slope is $b = 210 / 100 = 21 / 10$ and $a = 11.5 - (21 / 10) (5) = 1$ , giving $\hat{r} (q) = 1 + (21 / 10) q$ .

3.1F1F2A1step 1.1step 2.1algebra∎

At the $q = 5$ check standard, the pooled line predicts $11.5$ , so its day-specific residuals are $11 - 11.5 = - 0.5$ and $12 - 11.5 = 0.5$ . Inverting the pooled line maps the two observed responses to $10 / (21 / 10) = 100 / 21$ and $11 / (21 / 10) = 110 / 21$ , whereas the separate day lines each map their own response back to $q = 5$ . The three-standard table exposes the differing slopes in this stipulated example; it supplies no empirical evidence, statistical test, or general justification for pooling calibration curves.

Remarks

There is one stipulated reading per standard per day, so the table does not estimate repeatability, a variance component, a control limit, or uncertainty. It only demonstrates the finite arithmetic of comparing three standard responses under two specified lines.

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

8 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources

NIST/SEMATECH Engineering Statistics Handbook, Chapter 2: Measurement Process Characterization (standard reference, not scraped)
JCGM 100:2008, Evaluation of Measurement Data — Guide to the Expression of Uncertainty in Measurement, Annex H.3.1–H.3.3 (standard reference, not scraped)
Fan Li, STA610 Lecture 6 Notes: Linear Mixed-Effects Models (standard reference, not scraped)