Machine Learning Approaches for Pharmaceutical Demand Forecasting: A Bibliometric and Systematic Review of Methods and Research Trends

Abstract

Pharmaceutical demand forecasting plays a critical role in ensuring drug availability, reducing waste, and improving supply chain efficiency in healthcare systems. In recent years, machine learning (ML) techniques have emerged as promising tools for addressing the complex and nonlinear characteristics of pharmaceutical demand. This study presents a systematic and bibliometric review of machine learning approaches applied to pharmaceutical demand forecasting. A structured literature search was conducted in the Scopus database using Boolean query strategies to identify relevant studies published over the past decade. The retrieved publications were screened according to predefined inclusion and exclusion criteria. Bibliometric analysis was performed using VOSviewer to map research trends, collaboration networks, and keyword co-occurrence patterns. The analysis reveals a rapidly growing body of research in this field, with algorithms such as long shortterm memory networks, random forest, artificial neural networks, extreme gradient boosting, and linear regression widely employed for pharmaceutical demand prediction. Compared with traditional statistical models, these machine learning techniques demonstrate superior capability in capturing nonlinear patterns, seasonality, and complex demand dynamics. The bibliometric results highlight several emerging research themes, including machine learning–driven supply chain management, sustainable pharmaceutical logistics, and hybrid forecasting models. In addition, the integration of advanced technologies such as block chain and enterprise resource planning systems is increasingly explored to improve transparency, data reliability, and forecasting performance. Overall, this review provides a comprehensive synthesis of current methodologies and research trends in pharmaceutical demand forecasting. The findings offer valuable insights for researchers and practitioners seeking to develop more accurate, scalable, and interpretable forecasting models for pharmaceutical supply chain management.