# Agent-Based and Compartmental Modeling for Scenario Literacy

## Status

Current repository artifact for IARPG-OPS-2 2.0.18-wip. Review status: **reviewed with limitations for scenario literacy and public explanation**. The preserved source file remains individually addressable as provenance. This report is not an official IARPA publication.

## Purpose

Translate the supplied epidemiological-modeling material into safe public guidance on model literacy, comparative scenarios, state machines, uncertainty, and validation.

## Scope

This canonical report covers the user-supplied file `Building Epidemiological Models.md`. It is authoritative for repository provenance, the bounded synthesis below, and IARPG content-routing decisions. It does not establish official agency affiliation, scientific model validity, public-health guidance, weapons effects, real-world geopolitical prediction, or a current software implementation.

## Executive Summary

The source describes browser-based agent models, compartmental models, finite-state transitions, numerical solvers, and visual teaching techniques. For IARPG, these ideas support explanation of local actor behavior, system-level outcomes, and side-by-side scenario comparison. The repository does not treat the source as medical guidance, an outbreak forecast, or a validated epidemiological model.

## Evidence Reviewed

- [Preserved source file](../../source-files/planetary-systems-and-scenario-modeling/Building%20Epidemiological%20Models.md)
- Source collection: `planetary-systems-and-scenario-modeling`
- Source SHA-256: `7f937df97e28cf1704f5cbd0d455ad6f8475e198d277359c39217ac14dd3a264`
- Source identity: user-supplied Markdown described as IARPA-related material; not authenticated as an official IARPA publication.
- [IARPA mission](https://www.iarpa.gov/who-we-are/history/our-mission) — official agency identity, research mission, and non-operational boundary (accessed 2026-07-22).
- [IARPA research programs](https://www.iarpa.gov/research-programs) — official program directory (accessed 2026-07-22).
- [GAO infectious disease modeling practices](https://www.gao.gov/assets/gao-20-372.pdf) — communication, model description, verification, and validation (accessed 2026-07-22).
- [NIST digital-twin credibility work](https://www.nist.gov/programs-projects/digital-twins-advanced-manufacturing) — verification, validation, and uncertainty considerations (accessed 2026-07-22).

## Reviewed Synthesis

### Publication Decision

Retain the source as preserved research evidence. Reuse only the bounded statements in this section and the repository-authored findings below. Do not treat the original title, acronym expansion, code, quantitative claims, named technology stack, or scenario outputs as current fact without a new review.

### Claim Dispositions

| Claim ID | Topic | Disposition | Current bounded statement |
|---|---|---|---|
| `PSR-OPS2-218-02-01` | agent and compartmental models | **retained-method-guidance** | Retain the distinction between local-agent simulation and aggregate compartmental flow models. |
| `PSR-OPS2-218-02-02` | public-health prediction | **rejected-for-public-reuse** | Do not present the source code or outputs as medical advice, outbreak prediction, or policy recommendation. |
| `PSR-OPS2-218-02-03` | causal communication | **retained-method-guidance** | Prefer parallel baseline and intervention scenarios with the changed assumptions stated explicitly. |
| `PSR-OPS2-218-02-04` | numerical implementation | **requires-domain-validation** | Treat numerical methods and parameter choices as illustrative until independently verified by qualified modelers. |
| `PSR-OPS2-218-02-05` | population profiling | **safety-restricted** | Do not infer individual dangerousness, compliance, health status, or identity from aggregate scenario models. |
| `PSR-OPS2-218-02-06` | uncertainty | **bounded** | Publish ranges, sensitivities, omissions, and validation limits beside any model output. |

### Attribution, Fairness, and Safety Boundary

IARPA is the **Intelligence Advanced Research Projects Activity**, a U.S. research organization that invests in high-risk/high-payoff research for Intelligence Community challenges and is not itself operationally focused. The phrase “Integrated Artificial Reality Planetary Atlas” is a fictional construction in the supplied material and must never be presented as the agency’s name, product, endorsement, or program. Apply equal evidence standards, disclose uncertainty, include rights and affected-community consequences, and keep all reuse non-actionable.

### Reuse Rule

Use the smallest applicable bounded statement above. Re-check current agency programs, laws, software, vendors, performance, market data, and scientific claims against current primary or authoritative sources before public factual reuse.

## Findings

- Agent-based models make local rules visible; compartmental models summarize population-level flows. Each reveals different information and hides different detail.
- State definitions and transition rules should be explicit, versioned, and inspectable before visual animation is treated as evidence.
- Side-by-side baseline and intervention scenarios communicate causality more honestly than changing several variables in one animated run.
- Sensitivity analysis and uncertainty ranges are more useful than one precise-looking output when parameters are weakly known.
- IARPG can adapt these lessons to fictional influence, logistics, trust, and institutional response while avoiding medical claims or real-person risk scoring.

## Decisions or Recommendations

- Route public readers to the [Scenario Modeling and Systems Thinking](/scenario-modeling) page rather than to implementation snippets in the preserved source.
- Label every model family with purpose, inputs, assumptions, state variables, outputs, uncertainty, verification, validation, affected communities, and prohibited interpretations.
- Prefer side-by-side scenario comparison, reversible state history, and accessible tables over spectacular but unexplained visualization.
- Keep real IARPA references confined to accurate, sourced descriptions of the agency and its public research programs.
- Preserve the original source unchanged in the source archive, but do not duplicate unsafe or misleading implementation detail into current public guidance.

## Risks and Limitations

- The preserved source contains unsupported, time-sensitive, implementation-specific, and sometimes hazardous detail.
- This review is editorial and architectural; it is not scientific peer review, model accreditation, public-health review, legal review, weapons-effects validation, or security certification.
- Official program pages can change after the research cutoff; future releases must re-check time-sensitive references.
- A scenario can improve reasoning while still embedding bad assumptions. Visible structure does not prove real-world validity.

## Validation Performed

- Verified the preserved source SHA-256 and one-to-one canonical mapping.
- Compared agency identity and selected program descriptions against official IARPA pages on 2026-07-22.
- Compared model-credibility language with GAO and NIST guidance.
- Reviewed the synthesis for attribution, safety, accessibility, international fairness, uncertainty, and non-prediction boundaries.
- Verified repository-relative links and required report sections through the local memory and source-integrity validators.

## Memory References

- [`.uai/planetary-systems.uai`](../../../.uai/planetary-systems.uai)
- [`.uai/simulation.uai`](../../../.uai/simulation.uai)
- [`.uai/scenario-comparison.uai`](../../../.uai/scenario-comparison.uai)
- [`.uai/representation-safety.uai`](../../../.uai/representation-safety.uai)

## Related Durable Documents

- [Planetary Systems Research Integration and Scenario Modeling Report](planetary-systems-research-integration-and-scenario-modeling-report.md#findings)
- [Rapid Deep Research Methodology and Guidance](rapid-deep-research-methodology-and-guidance.md#findings)
- [Equal Standards Across Unequal Records](equal-standards-across-unequal-records-international-institutional-fairness-rights-and-evidence-audit.md#mandatory-analytical-distinctions-and-rights-safeguards)

## Supersession Status

Current for IARPG-OPS-2 2.0.18-wip until superseded by a later reviewed synthesis. The preserved source remains archival provenance and does not override active `.uai` memory or verified implementation.