Intermittent Demand Forecasting for Spare Parts Using Artificial Neural Networks and Deep Learning: Literature Review

Forecasting Intermittent demand for spare parts is essential for enhancing inventory management, particularly in industries where unplanned equipment downtime and inventory holding costs are significant. Conventional forecasting methods often underperform in handling the sporadic and nonlinear natur...

Full description

Saved in:
Bibliographic Details
Main Authors: Omnia Nabil, Nahid Afia, T Ismail
Format: Article
Language:Arabic
Published: Assiut University, Faculty of Engineering 2025-11-01
Series:JES: Journal of Engineering Sciences
Subjects:
Online Access:https://jesaun.journals.ekb.eg/article_442296_000672de78d030af83e3fa88862fdee7.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Forecasting Intermittent demand for spare parts is essential for enhancing inventory management, particularly in industries where unplanned equipment downtime and inventory holding costs are significant. Conventional forecasting methods often underperform in handling the sporadic and nonlinear nature of intermittent demand. This paper presents a focused literature review on the use of Artificial Neural Networks (ANNs) and Deep Learning (DL) techniques for forecasting intermittent demand. Unlike general reviews that survey all forecasting approaches, this study concentrates specifically on neural and learning approaches to capture nonlinear patterns. The findings demonstrate that ANN and DL-based models generally outperform classical methods in forecasting accuracy, especially under highly irregular demand. Despite the advances, the availability and quality of datasets remain a significant limitation in developing robust models. Future research directions are identified, including the need for improved feature engineering, architecture optimization, and model interpretability. This review aims to support researchers understanding the potential and challenges of neural approaches for forecasting of intermittent demand for spare parts.
ISSN:1687-0530
2356-8550