Перейти к содержанию
Научная статья

Artificial intelligence-driven statistical analysis in customs authorities: advansed models, institutional constraints and strategic prospects

Аннотация

The exponential growth of cross-border trade has led to unprecedented increases in customs-related statistical data, necessitating advanced analytical tools beyond traditional econometric approaches. This study develops an integrated Artificial Intelligence (AI) framework for customs data analytics, combining machine learning, econometric modeling, and data mining techniques. A hybrid AI-econometric efficiency model is proposed to quantify the impact of AI adoption on operational performance. Using a simulated dataset reflecting customs operations, the study demonstrates that AI-driven systems improve fraud detection accuracy by up to 35%, reduce processing time by over 80%, and optimize revenue collection. The paper contributes to the literature by introducing a multi-layer intelligent customs analytics architecture, addressing institutional constraints, and proposing policy-relevant solutions for emerging economies.

Ключевые слова