Forecasting Semiconductor Raw Material Availability for Timely Production and Delivery Optimization 

Abstract

The semiconductor industry relies on a consistent and efficient supply of raw materials, such as wafers, to meet fluctuating production demands. However, inaccurate demand forecasting often leads to inefficiencies, including overstocking, which ties up capital and increases storage costs, or understocking, which disrupts production schedules and results in lost orders. Ensuring an optimal supply of materials is critical to maintaining production efficiency, reducing waste, and improving overall profitability. To address this challenge, this study develops a forecasting model using time series analysis to predict future demand for critical raw materials. The models evaluated include ARIMA (Autoregressive Integrated Moving Average), ARIMAX (Autoregressive Integrated Moving Average with Exogenous Variables), SARIMA (Seasonal Autoregressive Integrated Moving Average), and SARIMAX (Seasonal Autoregressive Integrated Moving Average with Exogenous Variables). The results indicate that the SARIMAX model outperforms the other methods by incorporating seasonal trends and external factors affecting demand, leading to more accurate predictions. This research aligns with the Sustainable Development Goals (SDGs), particularly SDG 9 (Industry, Innovation, and Infrastructure) and SDG 12 (Responsible Consumption and Production). By enhancing forecasting accuracy, semiconductor manufacturers can optimize resource allocation, minimize waste, and improve supply chain efficiency, contributing to more sustainable industrial operations. Additionally, better demand prediction ensures a stable supply of essential components, supporting technological advancements and infrastructure development. Implementing robust forecasting models enables semiconductor companies to enhance production continuity, reduce material waste, and promote sustainable manufacturing practices.