Integrasi Analitika Prediktif pada Model Newsvendor untuk Optimasi Persediaan: Kajian Literatur

Abstract

The Newsvendor model is one of the most widely used approaches for inventory decision-making under uncertain demand conditions. However, the conventional Newsvendor model has a fundamental limitation because it assumes that the demand distribution is known prior to making ordering decisions. The advancement of predictive analytics offers an opportunity to overcome this limitation by utilizing historical data and machine learning algorithms to support inventory decision-making. This study aims to review the development of predictive analytics integration into the Newsvendor model, identify the methods employed, and analyze

their contributions and implementation challenges in inventory optimization. This research adopts a literature review approach by examining scientific articles obtained from Google Scholar, ScienceDirect, SpringerLink, IEEE Xplore, and Scopus. The selected literature focuses on the Newsvendor model, predictive analytics, and inventory optimization, and is analyzed descriptively through identification, classification, and synthesis of previous studies. The results indicate that research on the Newsvendor model has evolved from probability distribution-based approaches to data-driven approaches. Various predictive analytics methods, including Linear Regression, Random Forest, Extreme Gradient Boosting (XGBoost), Artificial Neural Networks (ANN), Long Short-Term Memory (LSTM), and Deep Learning, have demonstrated the ability to improve demand forecasting accuracy and support more optimal ordering decisions. Furthermore, recent studies have shifted toward integrating forecasting and optimization through forecasting–optimization, data-driven newsvendor, and end-to-end learning approaches. Nevertheless, the implementation of these approaches still faces several challenges, including the need for high-quality historical data, computational complexity, and the limited interpretability of certain machine learning and deep learning models. This literature review is expected to provide valuable insights for researchers and practitioners in understanding the development of predictive analytics applications in the Newsvendor model and to serve as a foundation for future research on data-driven inventory optimization.

Keywords: Newsvendor; Predictive Analytics; Machine Learning; Inventory Optimization; Literature Review.