La Gestión organizacional: Detección de objetos, una perspectiva desde la gestión para la eficiencia económica y la automatización de decisiones

Abstract

In a global context where technology and innovation are strategic pillars for economic development and organizational efficiency, this study analyzed advanced object-detection methods and their impact on organizational management. Object detection, traditionally associated with security, is presented here as a key resource that optimizes processes, reduces operational costs, and supports managerial decision-making in areas such as logistics, quality control, and risk management. The methodology employed followed the SEMMA process model (Sample, Explore, Modify, Model, Assess), consisting of five stages. Within this framework, a deep-learning approach was implemented using neural networks to evaluate accuracy and processing speed through the You Only Look Once (YOLO) algorithm, which enabled real-time detection by segmenting images into grids with simultaneous predictions of objects and classes, all executed in Jupyter Notebook using the Python programming language.

A diverse dataset was used, including security-related objects (firearms, sharp weapons) and common objects (animals, fruits, vehicles), demonstrating applicability across different economic contexts. This supports proactive management, optimal resource allocation, and increased competitiveness through dashboard-based monitoring. The results highlight the strategic value of object detection, achieving 92% accuracy and producing one analytical control dashboard as a driver of digital transformation, positioning this technique as an integrated solution capable of addressing contemporary challenges in economics and organizational management that addresses contemporary challenges in the economy and organizational management.