Document execution ownership and statistical integration qualification

CHANGELOG.md

@@ -4,6 +4,11 @@ All notable changes to this project will be documented in this file.
 
 ## [Unreleased]
 
+### Documentation
+
+- Record execution ownership, host-result boundaries, Fisher-screen integration
+  requirements and separate qualification targets for penalized stacking and EVSI.
+
 ## [0.6.0] - 2026-10-20
 
 ### Added

docs/guide/execution-roadmap.md

@@ -0,0 +1,77 @@
+# Execution and statistical integration plan
+
+Engineering decisions recorded on 2026-10-21. Keep NumPy as the core dependency
+and optional integrations behind extras. The shared
+[execution roadmap](https://github.com/ACCIDDA/op_engine/blob/main/docs/development/execution-roadmap.md)
+and [baseline protocol](https://github.com/ACCIDDA/op_engine/blob/main/docs/development/execution-baselines.md)
+describe the numerical packages and representative measurements.
+
+## Preserve package responsibilities
+
+trade-study owns generic replicate orchestration, scores, result tables,
+comparisons, optional surrogates and execution services. A simulator may use
+NumPy, JAX, Torch or CuPy internally while returning explicitly declared host
+results. This does not require making every runner/table operation GPU-native.
+Inference-specific priors, measurement models, sensitivities and posterior
+updating belong in OP Infer or a caller-supplied adapter.
+
+No Rust rewrite or full Julia port is committed by this plan. Existing Julia
+repositories are inactive design scaffolds. Native array transfer and host
+serialization boundaries must remain explicit. Generic runners should retain
+failures and stable task identities regardless of a simulator's backend.
+
+## Fisher feasibility screening
+
+A future Fisher screen can supply a scalar result to existing constraint and
+replication machinery. Consume an explicit information artifact or caller-owned
+screen instead of adding a compiler/inference dependency to the core. The
+artifact records parameter coordinates/scales, observation noise, design,
+nuisance treatment, reference parameter regime and numerical tolerances.
+
+Use measured-output parameter sensitivities, not the instantaneous RHS state
+Jacobian. Structural gauge freedoms need declared treatment; a known singular
+nominal Fisher matrix is not grounds for rejecting every otherwise useful design.
+Local numerical rank is a screen, not a structural-identifiability proof. Estimated
+screen uncertainty needs an actual sampling/replication contract before a
+probabilistic feasibility claim.
+
+## Predictive stacking and dimension penalties
+
+The existing Bayesian path delegates to ArviZ predictive stacking. Preserve that
+objective. Subtracting `IDOF * log(n)` from LOO scores is not an established
+correction and must not become an automatic behavior of `stack_bayesian`.
+Nominal parameter count, a reported gauge-quotient dimension, identifiable
+dimension and singular-model information penalties are distinct quantities.
+
+A proposed penalized alternative first needs a named objective, normalization,
+unit of comparison, definition of sample size, assumptions and independent
+qualification. Identical predictive distributions should give the same ensemble
+prediction under duplicate representations; their individual weights can be
+nonunique. Deduplication/grouping is a separate policy from complexity penalties.
+
+See [predictive stacking](https://arxiv.org/abs/1704.02030) and
+[singular-model information criteria](https://jmlr.org/papers/v14/watanabe13a.html).
+
+## Surrogate acquisition and EVSI
+
+Expected improvement over a current best score is a Bayesian-optimization
+acquisition target. Decision-theoretic expected value of sample information
+requires actions, utilities, a distribution of future observations and posterior
+updating before optimizing the action. A Gaussian surrogate alone does not make
+general EVSI analytic.
+
+Keep those APIs and evidence separate. A generic acquisition may consume a
+surrogate, while OP Infer or the caller supplies statistical updating and utility
+semantics for EVSI. The current RF tree spread is a relative disagreement measure,
+not calibrated posterior uncertainty. Qualification needs analytic decision
+fixtures, retained simulation failures and Monte Carlo error before claims about
+decision value. See the [EVSI formulation](https://doi.org/10.1137/19M1284981).
+
+## Sequence
+
+First reconcile overlapping issues and establish numerical-package baselines.
+Then integrate forward/provenance and shared statistical interfaces. Add Fisher
+screen adapters after sensitivity contracts are qualified. Keep penalized
+stacking and EVSI implementation gated on their separately specified targets.
+Forecasting consumers reuse trade-study execution and scoring rather than
+introducing a second replicate runner.

mkdocs.yml

@@ -63,6 +63,7 @@ nav:
   - Home: index.md
   - Capabilities and example checks: guide/capabilities.md
   - User Guide:
+      - Execution and Integration Plan: guide/execution-roadmap.md
       - Reactor Design (CSTR): guide/cstr.md
       - Hyperparameter Sweep (sklearn): guide/sklearn.md
       - Assay Costs and Annotations: guide/assay.md