Reproducible synthetic finite-world experiments for capability-preserving household payment transitions
Year: 2026
Type: dataset
Abstract: This research object contains synthetic data, executable Python source code, model specifications, independent verification procedures, results, and vector figures supporting the manuscript “Adaptive Security of Payment Systems: A Formal Test of Resilience under Cash Restrictions”. The study evaluates whether a candidate household payment portfolio preserves the mandatory and incumbent-feasible purchases of a reference portfolio under declared disruptions, travel constraints, purchase deadlines, payment-rail permissions, and limited information. The repository includes a finite-world nonanticipative-policy model, a 384-configuration sequential-resource diagnostic, 370 independent multistage policy comparisons, nine paired sufficiency/preservation cases, 176 exploratory scaling instances, and a 1,440-configuration time-critical synthetic pharmacy grid yielding 4,320 portfolio-feasibility decisions. It also includes separate euro-area same-currency mobility contrasts, clearly delimited derived aggregate evidence, figure-generation scripts, and bounded two-world TLA+/TLC model-checking artifacts (65 configured properties with expected results). All experiment-world counts arise from deliberately constructed, deterministic model inputs, not empirical household observations, transaction logs, representative surveys or estimates of actual merchant availability. Offline-payment feasibility is conditional on explicitly stated scheme, deployment, merchant, and completion assumptions. The bounded model checks do not establish universal policy correctness, real-world payment finality, or deployment readiness. Publicly reported aggregate figures are cited in source attribution files and must not be treated as observations of individual essential-purchase completion. The archive contains clearly organized studies/, evidence/, figures/ and verification/ directories; instructions for reproducing the computations, original scientific source and outputs, figure files, file-level licence boundaries, source attributions and SHA-256 integrity checks. The project-owned software is released under the MIT License, while the authors' original synthetic datasets, generated figures and explanatory documentation are provided under CC BY 4.0. Third-party source reports, raw survey data, and unpublished manuscript or editorial files are excluded.
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Keywords: synthetic data, reproducibility
Open Access status: green
OpenAlex ID: W7233981992
