Compliance Algorithm for Socio-Environmental Safeguards in Legal Amazon: A Traceability and Risk Assessment Model for Agricultural Commodities Chain under Regulation (EU) 2023/1115
Year: 2026
Type: article
Abstract: Regulation (EU) 2023/1115 requires agricultural commodities placed on the European market to be deforestation-free from 31 December 2020, produced in line with laws in its country of origin, and covered by a due diligence statement identifying production plots by geolocation. Demonstrating compliance depends on a unit of analysis—rural property and plot—that falls between municipal-scale trade-flow models and pixel-scale remote sensing products, while relying on heterogeneous and consistently incomplete databases. This article presents COMPLY-SOY-MA, a model for assessing traceability, risk, and socio-environmental compliance across agricultural commodities supply chains in Legal Amazon. Its main contribution is a credibility estimator that separates two often-confused dimensions: substantive compliance, measured through a composite index of thirty-one indicators aggregated using power means with controlled compensability, and supporting-evidence quality, measured through a five-indicator index. Risk is estimated by shrinking observed deficiency toward a territorial prior, weighted by the power of the evidence index, so opaque units are not assessed using averages from units that disclose information. Under isolated changes in evidence, providing more evidence reduces risk only for genuinely compliant units. Traceability is modelled as a mass-flow-weighted multigraph, using origin entropy and a formal distinction between exposure and contamination index. Scope is stated clearly: the study delivers a formal model and reference implementation, with results based on mathematical derivation and domain characterization. It specifies but does not execute a validation protocol, and its risk score is not a calibrated probability.
DOI: 10.66104/5gcafj90
Institutions: Universidade Federal do Maranhão, Universidade de Taubaté
Language: English
Subfield: Ecology
Field: Environmental Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7214225231