Cross-Border E-Commerce Demand Forecasting Model Based on Combining Time Series Analysis and Genetic Algorithm

In the context of globalization, cross-border e-commerce has developed rapidly, but demand is volatile, resulting in inefficient inventory management and supply chain. This paper aims to combine time series analysis with genetic algorithms to establish an accurate demand forecasting model for cross-border e-commerce to improve forecasting accuracy and optimize inventory management. Firstly, detailed weekly historical sales data are collected, and time series analysis methods are used to perform initial data conditioning, trend analysis, seasonal decomposition, and noise removal procedures; then, a forecasting model is established, ARIMA is applied to predict trends, and genetic algorithms are used to optimize parameters to enhance model performance. Extensive comparative experiments achieve a prediction accuracy of 99.4% with a root mean square error (RMSE) as low as 0.16, showing significant improvement. The combination of demand forecasting models, time series analysis, and genetic algorithms can substantially respond to changes in cross-border e-commerce demand, improve inventory management processes, and provide strong support for enterprise decision-making.