Ten documented projects.
StudyHow do food-loss assumptions change the picture of food availability?Food & resources · Historical modeled results
Food loss and availability
Purpose
We want to understand how assumptions about food loss change estimated availability and the ranking of food categories. The aim is to make those assumptions visible, so a model’s answer can be interpreted properly.
Method
Use documented USDA ERS inputs, define the loss scenarios, and compare the resulting rows and rankings with an independent checker.
Recorded evidence
The October 14 review records completed historical studies of 53,010 and 29,184 rows. A separate MIX study covered 6,840 rows and 96,139 checker comparisons, with zero discrepancies in that check.
What this does not establish
These are modeled availability scenarios. They do not measure actual household or national food waste, and agreement between checkers does not validate every real-world assumption.
Next step in the dated review
Resolve outstanding source-to-component definitions and denominators before considering further sensitivity studies.
Food-Loss Analysis · E32. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 2. Food-Loss Analysis E32. Source record retained privately.
StudyHow do household energy use and bills differ across income groups?Energy & households · Results with open discrepancies
Household energy and costs
Purpose
We want a clearer view of how households use energy and what they spend. Useful comparisons need consistent units, the right survey weights, and a clear account of what the data can support.
Method
Analyze the pinned EIA 2020 RECS household file, preserve the survey design, and reproduce the main comparisons independently.
Recorded evidence
The dated review records accepted historical analyses using a file with 18,496 households and 799 fields, including income-group means and component and sensitivity checks.
What this does not establish
There are 166 unresolved reconciliation failures. Income-burden claims are withheld because the income and expenditure time bases do not align. The accepted earlier studies do not resolve those issues.
Next step in the dated review
Investigate the discrepancies and complete the source and reporting repairs before making additional comparisons.
Household Energy Analysis · E24. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 3. Household Energy Analysis E24. Source record retained privately.
StudyDo PFAS treatment models remain consistent when we check their assumptions?Water & environment · Mixed simulation evidence
PFAS and treatment models
Purpose
We want to examine the evidence behind PFAS treatment-model claims. That means keeping reaction rates, concentration changes, treatment endpoints, and energy accounting distinct.
Method
Bind a model to its source equations and units, use finite simulations and synthetic fault checks, and compare the numerical results independently.
Recorded evidence
The October 14 review records an accepted E02 simulation grid of 625 vectors. E01 failed its coarse comparison: 40 of 48 comparisons failed. Both outcomes remain in the record.
What this does not establish
This work establishes no water-safety result, laboratory treatment performance, mineralization discovery, or confirmed error in a published paper. A simulation is not a physical water test.
Next step in the dated review
Reconcile the fitting failure and verify the original equations, volumes, endpoint definitions, and energy boundaries before considering further numerical work.
PFAS Clean-Water Research. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 9. PFAS Clean-Water Research. Source record retained privately.
StudyHow does travel behavior differ for people with disabilities?Access & communities · Methods preparation
Disability and accessible travel
Purpose
We want to understand aggregate differences in travel frequency, distance, and mode. The goal is evidence that can inform better questions about access and mobility without making assumptions about individuals.
Method
Prepare an NHTS survey analysis that respects household, person, trip, and diary-day records, with correct weights and uncertainty calculations.
Recorded evidence
Two covariance implementations each passed 32 synthetic cases. The dated review records no empirical travel result; production source bindings remain incomplete.
What this does not establish
Synthetic checks validate parts of the calculation. They do not establish a real-world disability-related travel difference or a causal explanation.
Next step in the dated review
Bind the source variables, survey design, missingness rules, and tested calculation methods before any empirical comparison.
Disability-Related Travel · E34. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 4. Disability-Related Travel E34. Source record retained privately.
StudyHow much of a household’s spending goes toward essentials?Energy & households · Source validation blocked
Essential household spending
Purpose
We want to compare essential-expense shares across U.S. income groups. The aim is a transparent comparison where readers can see which expenses count as essential and how each share was calculated.
Method
Use documented 2022 BLS Consumer Expenditure tables, source-cell references, three explicit baskets, and independent arithmetic checks.
Recorded evidence
The frozen question specifies 143 means, three baskets, and 33 ratios. The October 14 review records failed source-crop validation and no qualified spending comparison.
What this does not establish
No household-spending finding is established by this record. A working setup or a partially extracted table is not a validated comparison.
Next step in the dated review
Recover the source-validation diagnostics, repair the extraction under the existing research process, and independently reconcile all required source values.
Household Spending Comparison. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 1. Household Spending Comparison. Source record retained privately.
StudyWhat can county-level industry data tell us about regional economies?Access & communities · Historical aggregate results
Regional industry and concentration
Purpose
We want to understand how the scale and concentration of Texas industries differ by county and by measure. Coverage matters as much as a ranking: an incomplete county should remain visibly incomplete.
Method
Compare Census County Business Patterns measures on matched county-sector groups, retain suppressed or missing cells, and reproduce the rankings independently.
Recorded evidence
The dated review records a follow-on study with 762 profiles and 52 complete counties per measure. It preserves 202 incomplete counties without assigning them ranks.
What this does not establish
These are aggregate industry comparisons. They do not establish a company’s performance, creditworthiness, or financing eligibility. The review also notes that accepted follow-on results still needed workbook integration.
Next step in the dated review
Reconcile the accepted results into the canonical research workbook and verify source coverage before any additional time or measure comparison.
Texas B2B Industry Briefs. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 5. Texas B2B Industry Briefs. Source record retained privately.
StudyCan material-property predictions hold up under a fair test?Materials & prediction · Preparation unresolved
Trustworthy materials prediction
Purpose
We want to learn how reliably models predict material properties when the evaluation is protected from leakage. A useful result should survive a genuinely separate test, rather than benefit from information it should not have seen.
Method
Pin Matbench tasks and official evaluation folds, keep preprocessing within training data, and compare a fixed baseline with independently checked metrics.
Recorded evidence
The review records a defined research direction and incomplete preparation. It establishes no measured predictive accuracy or independent reproduction of a principal prediction result.
What this does not establish
Infrastructure checks are not prediction results. This project has not demonstrated a new material or a verified physical property.
Next step in the dated review
Resolve the preparation and capture failures, verify the data and fold identities, and qualify a simple baseline before testing a model comparison.
Materials Prediction · E15. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 6. Materials Prediction E15. Source record retained privately.
StudyCan we recover messages more reliably at the same communication cost?Communication & mathematics · Inconclusive simulation result
Reliable symbolic communication
Purpose
We want to compare symbolic coding schemes fairly: the same cost, the same simulated channel, and a clear account of errors and recovery. The aim is to learn when a proposed improvement actually meets its target.
Method
Freeze the channel and packet rules, charge for framing and feedback, check malformed controls, and reproduce the coding and cost calculations separately.
Recorded evidence
The project named Primes records a reproduction over 20,160 tuples and 144 cells. Its gain of 4.5714 percentage points fell below the frozen five-point threshold, so the result remained INCONCLUSIVE.
What this does not establish
This is a symbolic-channel simulation. It establishes no physical acoustic reliability, hearing, microphone, or field-channel result. The project name does not make it a prime-number theorem study.
Next step in the dated review
Finish the negative-control specification before considering any separately defined coding or repair experiment.
Primes · symbolic communication. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 10. Primes. Source record retained privately.
StudyWhat Goldbach patterns can we verify within a fixed finite range?Communication & mathematics · Historical finite verification
Finite Goldbach verification
Purpose
We want to compute and independently verify representations of even integers as pairs of primes, then examine clearly defined finite density patterns. The goal is a reproducible record of exactly what was checked.
Method
Use fixed input ranges, a separate primality and pair checker, explicit coverage rules, and versioned output records.
Recorded evidence
The dated review records witnesses for 49,999 even integers from four through 100,000. A smaller complete-pair domain contains 1,437,831 unordered pairs, with zero count-reproduction disagreements.
What this does not establish
Finite verification does not prove the universal Goldbach conjecture. The review’s accepted historical results do not imply that another computation is currently running.
Next step in the dated review
Complete the version-bound manifest and log provenance work for accepted outputs before considering any new finite study.
Goldbach Experiments. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 8. Goldbach Experiments. Source record retained privately.
StudyWhat happens when we test the Collatz map within explicit limits?Communication & mathematics · Primary study not run
Bounded Collatz experiments
Purpose
We want to examine finite integer trajectories while making the input range, transition counts, and stopping limits explicit. The value is a result that someone else can reproduce without guessing how the test was defined.
Method
Use exact integer arithmetic, independent fixtures and checking, and a predeclared domain of 9,968 seeds from 33 through 10,000.
Recorded evidence
The October 14 review records static and paper preparation. The proposed primary enumeration and its numerical quality checks remained unrun.
What this does not establish
There is no validated numerical result for that proposed primary study. Even a successful finite test would not prove the universal Collatz conjecture.
Next step in the dated review
Reconcile the frozen specification with the adapter and checker identities before any independently checked finite enumeration.
Collatz Experiments. Source: R640 Autonomous Operating Plan v1 — Problem Solver annex, 2026-10-14, 7. Collatz Experiments. Source record retained privately.
Seven proposed future studies.
FutureWeather observation gapsProposed future study · No completed result claimed
Proposed question
Do consecutive gaps distort temperature-extreme estimates more than scattered gaps, and can filling methods preserve honest uncertainty?
Purpose and limits
A proposed audit of how missing weather observations affect the metrics people use.
Proposal date: 2026-10-08. Source: Autonomous Research Portfolio and Pilot Decision, P2. No research work is started by this page.
FutureSimulated water-leak detectionProposed future study · No completed result claimed
Proposed question
Can calibration improve leak detection under demand shifts and sensor faults without increasing false alarms?
Purpose and limits
A proposed comparison on controlled leak and no-leak scenarios; field performance remains untested.
Proposal date: 2026-10-08. Source: Autonomous Research Portfolio and Pilot Decision, P3. No research work is started by this page.
FutureSolar estimate robustnessProposed future study · No completed result claimed
Proposed question
Which weather-input errors most change daily solar-energy estimates, and do uncertainty intervals cover the resulting errors?
Purpose and limits
A proposed sensitivity study of a documented photovoltaic model, rather than a savings or design guarantee.
Proposal date: 2026-10-08. Source: Autonomous Research Portfolio and Pilot Decision, P6. No research work is started by this page.
FutureStormwater intervention rankingsProposed future study · No completed result claimed
Proposed question
Do rankings of rain gardens, storage, and baseline drainage remain stable when rainfall and infiltration assumptions change?
Purpose and limits
A proposed simulation audit of ranking changes; it has not established flood-protection performance.
Proposal date: 2026-10-08. Source: Autonomous Research Portfolio and Pilot Decision, P7. No research work is started by this page.
FutureGeothermal fault mappingProposed future study · No completed result claimed
Proposed question
How well do lightweight fault-mapping methods generalize to held-out geographic regions?
Purpose and limits
A proposed spatial evaluation whose data access and practical fit remain unresolved.
Proposal date: 2026-10-08. Source: Autonomous Research Portfolio and Pilot Decision, P8. No research work is started by this page.
FutureAstronomical signal recoveryProposed future study · No completed result claimed
Proposed question
How do light-curve preparation choices affect recovery of injected small transit signals at a fixed false-positive rate?
Purpose and limits
A proposed TESS signal-recovery audit, with no planet discovery or confirmation claimed.
Proposal date: 2026-10-08. Source: Autonomous Research Portfolio and Pilot Decision, P9. No research work is started by this page.
FutureBattery-life generalizationProposed future study · No completed result claimed
Proposed question
Does apparent prediction accuracy hold up when entire battery cells or charging regimes are kept out of training?
Purpose and limits
A proposed evaluation of prediction reliability, rather than a safety or remaining-life guarantee for a real battery.
Proposal date: 2026-10-08. Source: Autonomous Research Portfolio and Pilot Decision, P10. No research work is started by this page.
No matching question. Clear the search to see the whole catalog.