Neural Network with Specialized Knowledge for Forecasting Intermittent Demand

Abstract

Demand forecasting is an essential part of an efficient inventory control system. However, when the demand has an intermittent or lumpy behavior, forecasting it becomes a challenging task. Several methods have been developed to solve this issue, but nonetheless, they only consider the information about the occurrence of demand, failing to assess the drivers of the data behavior. With the current digitalization of the industry, more data is available and, therefore, the chances of finding a causal relationship between the available data and the demand increases. Considering that, this paper proposes a single-hidden layer neural network for forecasting irregularly spaced time series with attributes conveying information about the past demand, seasonality of the data and specialized knowledge about the process. The neural network proposed is compared with benchmark neural networks and traditional forecasting methods for intermittent demand using three different performance measures on actual demand data from an industry operating in the aircraft maintenance sector. Statistical analysis is conducted on comparison results to identify significant differences in the forecasting methods according to each performance measure.

Improving demand forecasting for customers with missing downstream data in intermittent demand supply chains with supervised multivariate clustering

“…However, due to a lack of POS data, their study did not evaluate its results on the underlying unobserved demand data, instead using simulated data and out-of-sample delivery demand observations. Supervised machine learning approaches have been studied for intermittent demand forecasting such as neural networks (De Oliveira et al, 2020;Kourentzes, 2013;Lolli et al, 2017) and structured vector machines (Jiang et al, 2020). The advantage of these methods is that they do not rely on any underlying assumptions of the demand behavior.…”

“…Supervised machine learning approaches have been studied for intermittent demand forecasting such as neural networks (De Oliveira et al, 2020; Kourentzes, 2013; Lolli et al, 2017) and structured vector machines (Jiang et al, 2020). The advantage of these methods is that they do not rely on any underlying assumptions of the demand behavior.…”

“…This problem takes inspiration from similar problems faced in missing data interpolation, which have recently used improved machine learning techniques to great success (Che et al, 2018; Luo et al, 2018). Similar techniques have also been used in direct intermittent demand forecasting (De Oliveira et al, 2020; Jiang et al, 2020; Lolli et al, 2017). We extend these methods to the partial information challenge in a collaborative supply chain.…”

Previsão De Demanda Em Cadeias De Suprimentos: Uma Revisão Das Abordagens Baseadas Em Redes Neurais Artificiais

“…No entanto, sua aplicação enfrenta diversos desafios, a exemplo da disponibilidade de dados históricos precisos e confiáveis, uma vez que os dados de demanda podem ser afetados por diversos fatores externos, como condições climáticas e eventos políticos. Ademais, a escolha de arquiteturas de Redes Neurais (RN), funções de ativação e configurações de hiperparâmetros pode afetar significativamente a precisão das previsões (CHEN et al, 2018;DE OLIVEIRA et al, 2020;FANOODI;MALMIR;JAHANTIGH, 2019;WU et al, 2021WU et al, , 2023. Assim, surgem as seguintes Questões de Pesquisa (QP):…”