Algorithmic Legal Audit of Traffic Fine Payments

Arias R. · Lecture Notes in Networks and Systems (2026) · DOI: 10.1007/978-3-032-22236-7_25

Resumen

The study was grounded in the framework of administrative sanctioning law and municipal taxation, in dialogue with comparative literature where the regulatory architecture explained most observed outcomes. On that basis, it hypothesized that the typology set by Peru’s National Traffic Regulations determined the bulk of the variation in amounts paid, while district and month effects remained secondary; moreover, within each type, dispersion stayed moderate. The main objective was to assess, using computational methods and legal–tax criteria, the normative and revenue consistency of traffic-ticket payments in Lima during the first half of 2025. The central result confirmed the hypothesis: through a reproducible pipeline, a Random Forest achieved R2≈0.593 and MAE≈44,289 PEN; a log–Gaussian GLM posted a Cox–Snell pseudo R2≈0.916; and no outliers with |Z|≥3 were detected. It concluded that, in the period analyzed, payments exhibited internal proportionality and reflected materially uniform treatment across jurisdictions. As a key recommendation, the study proposed institutionalizing periodic audits with traceability and explainability—incorporating road-exposure indicators, records of discounts and aggravating factors, and local analyses such as SHAP—to reinforce the motivation of decisions, oversight, and continuous improvement of the sanctioning regime. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Instituciones: Universidad Autónoma del Perú