Box-Jenkins method for electricity consumption forecast for 2021-2023

• Autores: Anasco-Chata, Dennis Uriel, Huata-Panca, Percy, Mamani-Vilca, Ecler, Jimenez-Chura, Adolfo Carlos, Tapia-Catacora, Pablo Cesar, Añasco-Chata, Dennis Uriel
• Formato: artículo
• Fecha de Publicación: 2023
• Institución: Universidad Nacional Micaela Bastidas de Apurímac
• Repositorio: UNMB-Riqchary
• Lenguaje: español
• OAI Identifier: oai:revistas.unamba.edu.pe:article/28
• Enlace del recurso: https://revistas.unamba.edu.pe/index.php/riqchary/article/view/28
• Nivel de acceso: acceso abierto
• Materia: Box-Jenkins; Pronóstico; Energía eléctrica; Forecasting; Electric Power

The work was to determine a time series model based on the Box-Jenkins method, which is appropriately adjusted to the original series in the forecast of electricity consumption in the region of Puno Peru,  for the next three years (2021-2023), this would allow the company Electro Puno to have the model to support decision making, which conform to its corporate objectives in a medium-term time horizon, ensuring the distribution and marketing of electricity. The study was quantitative with non-experimental design, trend longitudinal cut, the population was 84 reports of monthly electricity consumption represented in (MWh/month), considering a commercial period of 7 years, the three phases of the Box-Jenkins methodology were applied as: identification, estimation and validation of the model, arriving to determine a multiplicative seasonal ARIMA (0,1,3)×(2,0,0,0)_12 model that adjusted perfectly to the series under study, being of "high precision" with a MAPE value of 2. 79% , finally in the last phase of the methodology, monthly forecasts of electricity consumption were made, with an average deviation for each forecast of a MAE value =625.80 MWh.