Demand Forecasting Model To Reduce The Mean Absolute Percentage Error By Applying Seasonal Breakdown Tools In A Sme In The Tourism Sector

• Autores: Ludeña Roman, Sayuri Arleth Renatta, Zelada Collazos, Sebastian, Corzo Chávez, Jorge Antonio
• Formato: objeto de conferencia
• Fecha de Publicación: 2024
• Institución: Universidad de Lima
• Repositorio: ULIMA-Institucional
• Lenguaje: inglés
• OAI Identifier: oai:repositorio.ulima.edu.pe:20.500.12724/21853
• Enlace del recurso: https://hdl.handle.net/20.500.12724/21853; https://doi.org/10.11159/icmie24.110
• Nivel de acceso: acceso abierto
• Materia: Pendiente

The research work is based on the analysis of demand in a tourism company using mathematical models. The methodology design presents a correlational and descriptive scope where the company's sales are collected to calculate the mean absolute percentage error in demand. With the help of machine learning tools, a predictive analysis will be carried out to estimate the sales for the following year, seeking to reduce the error using one of the selected mathematical models, calculate the necessary sales force, and thereby reduce the economic impact equivalent to $16 789,02. The MAPE (Mean Absolute Percentage Error) in the tourism sector is 12,03%. Through calculations using Python and RISK, a value of 15, 36% was obtained, reducing the MAPE by 4,24% compared to the year 2022. The Systematic Review of the Literature allows us to showcase the tools that can be developed in similar or atypical scenarios. The choice will depend on the behaviour pattern or trend. © 2024, Avestia Publishing. All rights reserved.