Supply chain forecasting model using neural networks

Overstock affects the supply chain, making it vulnerable and generating costs due to obsolescence. Up to now, it is needed to implement tools to forecast production demand with impact in supply chain inventory levels in order to know the reorder point and fulfill the client requirements. In this wor...

Descripción completa

Detalles Bibliográficos
Autores: Palafox-Palafox, Daniela, Medina-Marín, Joselito, Seck-Tuoh-Mora, Juan Carlos, Serna-Díaz, María Guadalupe, Hernández-Romero, Norberto
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2023
País:México
Institución:UNIVERSIDAD AUTÓNOMA DEL ESTADO DE HIDALGO
Repositorio:PÄDI Boletín Científico de Ciencias Básicas e Ingeniería del ICBI
Idioma:español
OAI Identifier:oai:repository.uaeh.edu.mx:article/11482
Acceso en línea:https://repository.uaeh.edu.mx/revistas/index.php/icbi/article/view/11482
Access Level:acceso abierto
Palabra clave:forecast
demand
supply chain
neural networks
simulation
pronóstico
demanda
cadena de suministro
redes neuronales
simulación
Descripción
Sumario:Overstock affects the supply chain, making it vulnerable and generating costs due to obsolescence. Up to now, it is needed to implement tools to forecast production demand with impact in supply chain inventory levels in order to know the reorder point and fulfill the client requirements. In this work, a simulation model of a supply chain was developed, considering four links (Supplier, Production, Distribution and Retailer) with their respective warehouses. Data concerned to inventory level and response time obtained from the simulation model were used for training of 100 different artificial neural network (ANNs) configurations, to identify the one with the best prediction for inventory levels. The ANN with the best performance (r2 = 0,99408, RMSE = 1,44217) contains 12 neurons in the input layer, 70 neurons in the first hidden layer, 60 in the second hidden layer and 4 neurons in the output layer.