Πώς η ανάπτυξη των μεγάλων γλωσσικών μοντέλων (LLMs) επηρεάζει τα ERP;
How does the development of Large Language Models (LLMs) affect ERP systems?
View/
Keywords
Συστήματα ERP ; Μεγάλα γλωσσικά μοντέλα ; Αυτόνομοι πράκτορες ; Εξόρυξη διαδικασιών ; Γνωστική αυτοματοποίησηAbstract
This bachelor thesis investigates the central research question of how the rapid development of Large Language Models (LLMs) and Generative Artificial Intelligence affects and transforms Enterprise Resource Planning (ERP) systems. The primary objective is to highlight the technological solutions AI provides to the inherent limitations of traditional ERPs, such as structural rigidity, complex graphical interfaces, and the inability to utilize unstructured data. To achieve this, a systematic literature review was conducted based on the PRISMA protocol, examining 105 contemporary academic and industrial sources from the period 2023-2026.
The study conducts an in-depth analysis of modern architectures, such as information retrieval from corporate documents (RAG) and interacting with the ERP via natural language (NL2SQL). Concurrently, it examines the practical integration of autonomous agents (Agentic AI) across critical enterprise modules (Finance, Supply Chain, Human Resources), as well as the catalytic impact of these models on Business Process Mining. The conclusions demonstrate that ERPs are definitively transitioning from passive data repositories to "cognitive" systems capable of active decision-making. However, it is established that their secure adoption necessitates the rigorous management of technological challenges, such as model inaccuracies (hallucinations), cybersecurity risks, and European regulatory compliance.