THE APPLICATION OF TIME SERIES ANALYSIS TO PREDICT MARKET TRENDS AND OPTIMIZE SUPPLY CHAIN PLANNING

This research investigates the application of time series analysis in predicting market trends and optimizing supply chain planning. The study aims to assess the effectiveness of predictive models, evaluate their impact on supply chain optimization, and identify best practices in their implementation. The methodology integrates both qualitative and quantitative approaches, utilizing ARIMA, exponential smoothing, and machine learning models such as Long Short-Term Memory (LSTM) networks. Data from 2020 to 2024 was analyzed using statistical tools, including Python and R, to validate model accuracy. The findings reveal that LSTM models achieved the highest predictive accuracy, with a Mean Absolute Error (MAE) decreasing from 1.9 in 2020 to 1.4 in 2024, outperforming traditional time series models. The correlation between market trend predictions and supply chain performance increased from 0.65 in 2020 to 0.85 in 2024, highlighting the growing effectiveness of predictive analytics. Linear regression analysis confirmed a significant reduction in inventory costs (slope = 1.26, R² = 1.00), emphasizing the financial benefits of predictive modeling. Based on these results, the study recommends integrating hybrid forecasting models, enhancing data governance, and investing in AI-driven analytics to optimize supply chain planning. Future research should focus on real-time data integration and stress-testing methodologies to further improve resilience against supply chain disruptions.