engineering.uncertainty.precision_mean

Three distinct suites: educational examples on the calculator page, this legacy regression count (9/9 golden + boundary), and the CVP verification suite below. They are not the same number. Property, metamorphic, round-trip, O3, and interface statuses/checks are recorded in the CVP Manifest . SKIP (0/0) means that suite has no published vectors.

This calculator CURRENT · Public schema 1.0.1 matches · Semantic contract ✓ · Production attested · Public/cache ✓ · Origin ✓

CVP verification suite

Protocol checks from the Evidence Manifest. Distinct from the legacy regression suite above. Verification revision 2026-09-23.o2-o3. Calculation version 1.0.1. CVP 1.0.0-proposed. 4/4 golden · 5/5 CVP boundary · 6/6 invalid · 1/1 property · 1/1 metamorphic · 8/8 O3 · 4/4 cross-interface · 6/6 CVP contract. digest 0f27e2e3a0f3. O2: ISO 21748 u = √(s_L² + s_r²/n) vs a separate-module identity (≤2 ULP). Not R = 2.8·s_R, not Type A from a sample, and not U = k·u. ≤2 ULP vs O3 applies only to the published tabulated ISO 21748 vectors (both components, repeatability only, laboratory only, an alias, and one awkward mean). It is not a whole-domain guarantee and does not cover R = 2.8·s_R, Type A from a sample, or U = k·u.

• Golden (oracle-backed) 4 / 4
• CVP boundary 5 / 5
• Invalid input 6 / 6
• Property 1 / 1
• Metamorphic 1 / 1
• O3 high-precision 8 / 8
• Cross-interface 4 / 4
• CVP contract 6 / 6
• Interfaces PASS · UI (SSR) / REST / MCP
• Interface claim ui-ssr is query-result HTML, not a live browser session. Error-path engine·REST·MCP 1/1 (status, code, calculation_version). SSR compared on URL-canonical requested calculations; empty query is idle (not an error) and JSON-typed object/array inputs are REST/MCP-only.
• Verified scope ISO 21748 u = √(s_L² + s_r²/n) for a mean of n≥2 replicates from stated s_r and s_L, + O3 mpmath tabulated u.
• Assurance Engineering
• Deferred Not R = 2.8·s_R, not Type A from a sample, and not U = k·u.

• release artifacts aligned PASS
• schema output required declared PASS
• schema valid input executes PASS
• engine success output validates schema PASS
• schema rejects incomplete input PASS
• rest success payload validates schema PASS

Fingerprint sha256:a67d5572dc7ae0ea43488504c7fd1d09d00f38697ef81d291ac70a4eac7b8eb0

Input schema sha256:cdbf42690f6344f56212f6f86accd1ae12e1b6299530d706be92f4482531fa22 · Output schema sha256:5f5d57b20a3a795bbd3651f849cb4e0ebf55a5844ae803071e46f1d7f2f1a33c · Build-time SHA-256 digests prove published artifacts agree with each other (release_integrity.scope=workspace). Production attestation GETs canonical URLs twice (public/cached view and origin), including Capability HTML, via npm run attest:production --write (published at /.well-known/calculatorx-production-attestation.json) and fails if the public Capability page advertises a different calculation_version. This is artifact consistency, not an independent immutable release log.

Tool

precision-uncertainty

Engine

precision-uncertainty

Engine build

2026-09-23

Evidence core

2026-09-23T12:37:01.564Z

Rendered

2026-10-22T08:54:34.789Z

Last reviewed

—

Assumptions

• ISO 21748: u = √(s_L² + s_r²/n) of a mean of n replicates from stated precision components. n≥2.
• s_r and s_L are ≥ 0. Both zero is valid (u = 0).
• Not R = 2.8·s_R, not Type A from a sample, and not U = k·u.

Limitations

• Takes stated s_r, s_L, and n≥2 — not a sample of observations
• n=1 is rejected so this cannot wrap s_R
• Not ISO 5725 R = 2.8·s_R and not Type A u_A = s/√n from values
• Does not compute expanded U = k·u