# Proposed 24-month research programme — implementation appendix

This appendix preserves the existing collaboration proposal and adds decision gates from the supplied development plan. It is a proposal, not a record of completed months, secured funding, laboratory access, institutional endorsement or physical validation. Facility experiments remain under the host institution’s approved procedures. Software interfaces and synthetic fixtures cannot replace those dependencies.

## Existing four phases retained

| Period | Existing phase | Existing scope |
| --- | --- | --- |
| Months 1–6 | Scope and review | Agree the research scope, review the models, define evaluation criteria and prepare funding proposals. |
| Months 7–12 | Build and first measurements | Develop the experimental setup and collect initial prototype measurements. |
| Months 13–18 | Refine and evaluate independently | Refine the approach, assess repeatability and arrange independent evaluation. |
| Months 19–24 | Consolidate and define the next stage | Consolidate the findings, prepare appropriate research outputs and define the next development stage. |

## Detailed milestone and decision-gate proposal

| Period | Work | Deliverable | Proceed only if |
| --- | --- | --- | --- |
| Month 1 | Repair release/domain/status defects; select one flagship primary endpoint | Corrected release, regression evidence, concise problem statement | Another person can run the check and understand one result without interpretation errors |
| Months 2–3 | Form host/data relationship; establish contracts and image/electrical ingestion | Partner work-package draft, data/rights plan, typed schemas, existing-model baselines | Actual data or a credible scheduled acquisition path exists |
| Months 4–6 | Calibrated pilot analysis; connected power/particle records; formal evaluation design | Pilot dataset, frozen baseline, reproducible package, funding-ready work plan | Calibration is adequate and the proposed endpoint is identifiable/measurable |
| Months 7–9 | Build the first measurement workflow with the facility; implement the chosen robust analysis | Initial non-fusion measurement records and comparison plots | Readings have traceable uncertainty and the model boundary is clear |
| Months 10–12 | Complete first held-out validation and reference-model comparison | First validated subsystem result or documented model failure | Improvement is demonstrated against a baseline; unresolved effects are listed |
| Months 13–15 | Refine the strongest method; add correlated uncertainty and a second campaign | Prediction-interval evaluation and independent session results | Method remains useful when acquisition conditions change |
| Months 16–18 | External rerun or second-site test; limited experiment-selection study | Independent evaluation report and preregistered selection comparison | A partner can reproduce the key result without the original developer operating the workflow |
| Months 19–21 | Consolidate source/transport/detector/electrical links where validated | Integrated research package, methods manuscript/data release where permitted | Every headline is traceable to its supported domain and evidence |
| Months 22–24 | Final independent review, adoption testing, follow-on proposal | Public/controlled release, documented limits, next-stage decision | Follow-on scope is justified by evidence rather than visual ambition |

### 13.2 Stop or narrow the project when needed

- **No usable data by Month 3:** focus the pilot on source/solver verification and data infrastructure; do not promise a learned detector model trained on unavailable data.
- **No independent response calibration:** stop species/energy inference and publish validated morphology/counting results instead.
- **The baseline already performs adequately:** investigate reliability, calibration transfer or workflow effort; do not add model complexity merely to claim AI.
- **Persistent unexplained collector residuals:** localize the discrepancy before optimizing the apparatus on a wrong model.
- **No external adoption or reviewer by Month 12:** reduce breadth and make one package easier to reproduce rather than opening more labs.
- **Scaling changes the governing regime:** record the validated small-scale result and create a separate next-stage physics question.

### 13.3 Team and budget assumptions

A serious 24-month programme spans software, data analysis and experimental expertise. It is not realistically represented by one developer adding interface controls. Institutional contributions may be in-kind, but their effort and facility costs should still be visible.

The following is **illustrative planning arithmetic**, not salary research, a vendor quote, an eligible-cost determination, or an expected grant award:

| Resource | Planning assumption | Cost |
|---|---:|---:|
| Scientific software engineering | 24 person-months × €8,000 loaded/month | €192,000 |
| ML/statistics/measurement research | 18 person-months × €9,000 | €162,000 |
| Instrumentation/experimental development | 12 person-months × €9,000 | €108,000 |
| Senior scientific review and independent evaluation | 6 person-months × €10,000 | €60,000 |
| Facility access and external calibration | Placeholder pending quotations | €50,000 |
| Equipment/data-acquisition contribution | Placeholder pending facility design | €40,000 |
| Compute, software access and archiving | Placeholder | €10,000 |
| Subtotal | 60 person-months plus listed costs | €622,000 |
| Planning contingency | 15% of subtotal | €93,300 |
| **Illustrative total** | 24-month multidisciplinary programme | **€715,300** |

This is not the cost of a fusion reactor, a power plant, or a new beam facility. A host may supply equipment, existing data, supervision or staff effort, changing the cash requirement. Avoid counting those contributions both as direct expenditure and as free capacity.

For an initial software/data pilot, scope roughly 6–12 person-months and price the actual facility/data requirements separately. Using a purely illustrative €6,000–€10,000 loaded person-month gives €36,000–€120,000 in effort before facilities and other costs. A small transfer voucher could support one external validation task within such a pilot; it should not be represented as funding the whole programme.

The figures above are retained planning arithmetic from the supplied proposal. They are not new salary research, procurement quotations, approved eligible costs, funding offers or a fusion-reactor budget. No external message or application was sent in this software work.

## First 30 days: proposed work sequence

| Days | Action | Definition of done |
| --- | --- | --- |
| 1–5 | Reproduce and repair B01–B04 in the source repository | Matching-release hosting check, correct inclusive temperature boundary, consistent run/export metadata and status |
| 6–10 | Promote heating closure; fix comparator units and stage labels | One result view shows electrical boundary, required heating and assumptions; units are explicit |
| 11–15 | Choose AlphaTrack or measured collector response as the primary pilot | One-page hypothesis, baseline, observable, necessary data and failure criterion |
| 16–20 | Produce a sample typed package and a short independent replay | A second person can reproduce the selected numerical/analysis output from documented inputs |
| 21–25 | Prepare the host/data work package | Named expertise needed, data/calibration list, contribution table, effort and permitted outputs |
| 26–30 | Review with a prospective scientific host and refine the scope | Documented feedback and a concrete next experiment or data task; no implied endorsement |

Current software implementation status is in `v750-requirements.json`. Its statuses do not establish that an independent person, institution or facility completed these proposed milestones. Keep research endpoints tied to actual data, preregistration and external review.