Demand forecasting is a critical component of supply chain management, essential for optimizing inventory levels, reducing costs, and enhancing customer satisfaction .This study tries to enhance the efficiency of supply chain by accurate demand forecasting using the deep learning models. This thesis investigates the application of deep learning models, particularly Long Short-Term Memory (LSTM) networks, to improve the accuracy of demand predictions. Traditional statistical method, Exponential Smoothing, was compared to advanced deep learning models, single layer and multi layer LSTM architectures. The hybrid model, which combines Exponential Smoothing with LSTM, which showed better performance than traditional approaches and the single layer LSTM. Despite initial expectations, the hybrid model did not outperform the multi layer LSTM probably because of weak performance of Exponential Smoothing. The findings underscore the potential of deep learning techniques in enhancing demand forecasting accuracy, offering valuable insights for data-driven decision-making in supply chain management.
Demand forecasting is a critical component of supply chain management, essential for optimizing inventory levels, reducing costs, and enhancing customer satisfaction .This study tries to enhance the efficiency of supply chain by accurate demand forecasting using the deep learning models. This thesis investigates the application of deep learning models, particularly Long Short-Term Memory (LSTM) networks, to improve the accuracy of demand predictions. Traditional statistical method, Exponential Smoothing, was compared to advanced deep learning models, single layer and multi layer LSTM architectures. The hybrid model, which combines Exponential Smoothing with LSTM, which showed better performance than traditional approaches and the single layer LSTM. Despite initial expectations, the hybrid model did not outperform the multi layer LSTM probably because of weak performance of Exponential Smoothing. The findings underscore the potential of deep learning techniques in enhancing demand forecasting accuracy, offering valuable insights for data-driven decision-making in supply chain management.
Enhancing Supply Chain Efficiency: Demand Forecasting Using Deep Learning Techniques
AHMADIESLAMLOO, ELAHEH
2023/2024
Abstract
Demand forecasting is a critical component of supply chain management, essential for optimizing inventory levels, reducing costs, and enhancing customer satisfaction .This study tries to enhance the efficiency of supply chain by accurate demand forecasting using the deep learning models. This thesis investigates the application of deep learning models, particularly Long Short-Term Memory (LSTM) networks, to improve the accuracy of demand predictions. Traditional statistical method, Exponential Smoothing, was compared to advanced deep learning models, single layer and multi layer LSTM architectures. The hybrid model, which combines Exponential Smoothing with LSTM, which showed better performance than traditional approaches and the single layer LSTM. Despite initial expectations, the hybrid model did not outperform the multi layer LSTM probably because of weak performance of Exponential Smoothing. The findings underscore the potential of deep learning techniques in enhancing demand forecasting accuracy, offering valuable insights for data-driven decision-making in supply chain management.File | Dimensione | Formato | |
---|---|---|---|
Ahmadieslamloo_Elaheh.pdf
accesso riservato
Dimensione
5.37 MB
Formato
Adobe PDF
|
5.37 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/78374