Thesis Desarrollo de un plan de negocio para EYON: un servicio basado en inteligencia artificial para reducir hurtos en cadenas de comercio minorista
Loading...
Date
2026-08
Journal Title
Journal ISSN
Volume Title
Program
Ingeniería Civil Informática
Departament
Campus
Campus Casa Central Valparaíso
Abstract
Esta memoria desarrolla un plan de negocio para EYON, un servicio SaaS B2B basado en inteligencia artificial, orientado a apoyar la reducción de hurtos hormiga en supermercados y cadenas de retail mediante reconocimiento de actividad humana, procesamiento local en el borde y privacidad por diseño. El estudio corresponde a una investigación descriptivo-propositiva con enfoque mixto, sustentada en fuentes secundarias y herramientas como Business Model Canvas, segmentación TAM/SAM/SOM, diseño conceptual de una arquitectura edge-cloud y evaluación financiera mediante VAN, TIR, ROI, payback y punto de equilibrio. El modelo proyecta 45 nodos activos al mes 36 y 131 al mes 60. Aunque los indicadores permanecen negativos al mes 36, al finalizar el horizonte se obtiene un VAN de $252.414.917, una TIR mensual de 3,194%, un ROI de 84,446% y un payback en el mes 52. El análisis de sensibilidad identifica la tarifa mensual y la velocidad de adopción como las variables más críticas. Debido a que no se realizaron pruebas en terreno, estos resultados representan una viabilidad conceptual sujeta a validación técnica y comercial. La principal contribución del trabajo es estructurar una base integral para el desarrollo, pilotaje y evaluación posterior de EYON en el contexto del retail chileno.
This thesis develops a business plan for EYON, an artificial intelligence-based B2B SaaS service designed to support the reduction of shoplifting in supermarkets and retail chains through human activity recognition, local edge processing, and privacy by design. The study follows a descriptive-propositional mixed-method approach based on secondary sources and tools such as the Business Model Canvas, TAM/SAM/SOM market segmentation, the conceptual design of an edge-cloud architecture, and financial assessment through Net Present Value, Internal Rate of Return, Return on Investment, payback period, and break-even analysis. The model projects 45 active nodes by month 36 and 131 by month 60. Although the financial indicators remain negative at month 36, by the end of the evaluation horizon the project reaches an NPV of CLP 252,414,917, a monthly IRR of 3.194%, an ROI of 84.446%, and payback in month 52. Sensitivity analysis identifies the monthly fee and adoption rate as the most critical variables. Since no field testing was conducted, these results represent conceptual feasibility subject to technical and commercial validation. The main contribution of this work is an integrated foundation for the future development, pilot testing, and evaluation of EYON within the Chilean retail context.
This thesis develops a business plan for EYON, an artificial intelligence-based B2B SaaS service designed to support the reduction of shoplifting in supermarkets and retail chains through human activity recognition, local edge processing, and privacy by design. The study follows a descriptive-propositional mixed-method approach based on secondary sources and tools such as the Business Model Canvas, TAM/SAM/SOM market segmentation, the conceptual design of an edge-cloud architecture, and financial assessment through Net Present Value, Internal Rate of Return, Return on Investment, payback period, and break-even analysis. The model projects 45 active nodes by month 36 and 131 by month 60. Although the financial indicators remain negative at month 36, by the end of the evaluation horizon the project reaches an NPV of CLP 252,414,917, a monthly IRR of 3.194%, an ROI of 84.446%, and payback in month 52. Sensitivity analysis identifies the monthly fee and adoption rate as the most critical variables. Since no field testing was conducted, these results represent conceptual feasibility subject to technical and commercial validation. The main contribution of this work is an integrated foundation for the future development, pilot testing, and evaluation of EYON within the Chilean retail context.
Description
Keywords
Hurtos hormiga, Inteligencia artificial, Reconocimiento de actividad humana, Retail, Modelo de negocio B2B SaaS, Privacidad por diseño
