Sarcouncil Journal of Applied Sciences Aims & Scope

Abstract: Inventory management plays a vital role in ensuring the efficiency of supply chains, where accurate demand forecasting is central to minimizing stockouts, reducing holding costs, and improving overall operational performance. Time series forecasting has emerged as a powerful approach for predicting future inventory requirements by capturing trends, seasonality, and irregular demand patterns. This review paper provides a comprehensive overview of traditional statistical methods such as ARIMA, SARIMA, and exponential smoothing, alongside advanced machine learning and deep learning models, including LSTM, Prophet, and XGBoost, with a focus on their applications in inventory management. Special attention is given to the use of R and Python, two widely adopted programming environments, which offer rich ecosystems of libraries and packages that facilitate model building, evaluation, and visualization. Through a critical analysis of existing literature and comparative implementations, this paper highlights the strengths, limitations, and applicability of various forecasting methods in real-world inventory contexts.