# Validation report — PE-Sim Explorer v1.2.0

Date: 04 September 2026

## Static integrity checks

- JavaScript syntax: PASS (`node --check app.js`)
- `index.html` static serving: PASS
- All JavaScript `$('<id>')` element references present in HTML: PASS
- External HTTP/HTTPS resources in runtime HTML/CSS/JS: NONE
- Backend/API dependency: NONE
- Default seed: 1729
- Metric bars: native HTML progress elements with `max=1` and concordance as `value`, explicitly labelled on a 0.00–1.00 scale
- CSP compatibility: PASS; no runtime inline style attributes are required for bar widths
- Reset behavior: restores benchmark-backbone controls and clears stale run output

## Deterministic engine smoke test

Test configuration: benchmark-backbone exploratory preset, mixed A1–A10 population, 15 research-active years, citation ecology 1.0×, N=200, seed=1729, Pareto S/C/P/I reference.

Observed smoke-test output:

- Run ID: `PESX-7339793a`
- PE concordance: 0.932608
- h-index concordance: 0.872774
- PE − h-index: +0.059834
- Comparable Pareto pairs: 10,387
- Researcher-type rows generated: 10/10 archetypes

These values are only a software smoke test and are **not** the frozen benchmark result.

## Researcher-type summary smoke test

For the same N=200 mixed run, the application successfully recalculated within-archetype concordance for all generated A1–A10 groups and returned type-specific best metrics, PE/h-index concordance, PE−h difference, medians and comparable-pair counts. At this smaller test size some archetypes had fewer than 100 within-type comparable pairs, which is why the interface explicitly reports pair counts.

## Non-advocacy / failure-mode smoke test

High-output / low-impact preset, A10-only population, N=600, seed=1729:

- Publication-count concordance: 0.9546
- PE concordance: 0.9545

This confirms that the Explorer can return a non-PE metric as the top performer under an adversarial scenario; the code does not impose PE as the winner.

## Frozen benchmark reference

The application displays but does not recompute or overwrite the frozen ICMR-PE-SIM v1.0.1 primary benchmark:

- PE Pareto concordance: 0.957384
- h-index concordance: 0.917406
- Difference: +0.039982
- Empirical 95% interval: [+0.036527, +0.044771]
- Strict PE-h discordant pairs: 7,092,548

## Scope limitation

The browser Explorer is intentionally smaller and interactive. It does not substitute for the frozen 200-population, 1.2-million-researcher-realization benchmark workflow or its complete Monte Carlo uncertainty and audit machinery.