Η συμβολή της τεχνητής νοημοσύνης στην βελτιστοποίηση της εφοδιαστικής αλυσίδας με έμφαση στην πρόβλεψη ζήτησης και τη διαχείριση αποθεμάτων
The contribution of Artificial Intelligence to the optimization of the supply chain with emphasis on demand forecasting and inventory management
View/
Keywords
Artificial IntelligenceAbstract
In the modern business environment, the inability to make reliable forecasts often leads either to overstocking, resulting in increased storage costs and capital commitment, or to product shortages, which directly impact the level of customer service and the company’s reputation. Thus, demand forecasting accuracy is undoubtedly a crucial success factor for businesses.
Traditional demand forecasting methods in supply chain management are rather limited, as they are mainly based on statistical techniques and historical data. They are unable to effectively incorporate external and non-linear factors. This, combined with poor data quality, restricts the ability to make timely decisions, creating inefficiencies throughout the entire supply chain and its individual stages.
Undoubtedly, Artificial Intelligence (AI) plays a decisive role in modern supply chain management. Specifically, in the transition toward modern demand forecasting and inventory management systems, a wide range of techniques is utilized, such as time series models, neural networks, Support Vector Machines, and other machine learning algorithms. These methods enable not only the detection of complex, non-linear relationships within large and heterogeneous datasets, but also self-improvement through iterative learning, dynamically leveraging market changes and providing higher forecasting accuracy.
Equally important for the smooth integration of AI methods into the supply chain is the incorporation of alternative data sources, such as social media, through Natural Language Processing and sentiment analysis techniques. In this way, the analysis of unstructured data captures consumer sentiment in real time with high accuracy. Additionally, blockchain technology ensures the exchange of reliable data among all participants in the supply chain, reducing information asymmetry and strengthening trust. Access to up-to-date and accurate real-time data improves the quality of inputs to predictive models, leading to more reliable forecasts and more effective inventory management, as well as improved transparency toward consumers.
This master’s thesis thoroughly examines the role of Artificial Intelligence in modern supply chain management, focusing on demand forecasting and inventory optimization. It is emphasized that through investments in digital infrastructure and the development of appropriate skills, AI can become an integral part of both modern and future supply chains, contributing to the creation of flexible and efficient systems, improving forecasting accuracy, reducing costs, and enhancing competitiveness.