Plan de mejora para la optimización de la revisión de declaraciones de importación mediante inteligencia artificial en Motorysa

Abstract

This report documents the work carried out during the professional internship in the Foreign Trade Department of Motorysa, a company with more than 55 years of experience in the Colombian automotive sector and the exclusive distributor of Mitsubishi and BYD in the country (Motorysa, n.d.). The internship was developed in support of the BYD and Mitsubishi vehicle lines, within the framework of the vehicle customs clearance process carried out in collaboration with the customs agency Almaviva S.A. It was identified that the manual review of import declarations, both before and after their official acceptance by the National Tax and Customs Directorate (DIAN), took an average of two and a half working days (approximately 20 hours) for a shipment of 1,500 vehicles. This process was also exposed to errors resulting from employee fatigue when manually comparing each field of the declaration with its supporting documents. As an improvement plan, a document validation matrix was designed and its use was implemented with the support of generative artificial intelligence tools, specifically Microsoft Copilot and Google Gemini. This made it possible to reduce the review time for the same volume of declarations to approximately ten minutes, representing a reduction of more than 99% in the time required. The results demonstrate the potential of artificial intelligence as a support tool for professional judgment in foreign trade processes. Therefore, it is recommended that the company’s Information Technology Department develop a specialized solution to ensure the continuity and further development of the proposed mechanism.