Propuesta metodológica de la planeación de la demanda y abastecimiento mediante inteligencia artificial para la empresa Autoradiador S.A.

Abstract

objective of this research is to optimize the demand planning and procurement process at Autoradiador S.A., Through the design of a methodological proposal based on Generative Artificial intelligence to mitigate operational inefficiency and purchasing subjectivity. The applied methodology employed a quantitative approach with the experimental scope, developing a functional prototype that integrates Medium-Code tools (n8n) and Large Language Models (LLMs) under a hybrid architecture. For validation, the algorithm’s results were constrated against the Category’s expert criteria using a representative sample of 67 SKUs. The main conclusions demonstrate that the system acts effectively as a Decision Support System (DSS), achieving an Inventory Adjustment Precision (PAI) of 94.3% and eliminating the risk of stockouts. Furthermore, high operational efficiency was verified, reducing execution times by 94.7% and decreasing process costs by 97.4% with an audited value of $0.30 USD per execution. These findings confirm that IA-driven automation is technically and financially viable, allowing the organization to standardize its logistical criteria and reorient resources towards commercial innovation.