Will AI Streamline ERP System Implementations in Enterprises?

ERP system implementation projects in enterprises are inherently complex and often unsuccessful due to the time-consuming and expert-dependent phases of requirements elicitation, analysis, and parameterization. The specific objective of this study is to evaluate the feasibility of leveraging large language models (LLMs) to automate these initial stages—specifically, to classify standard and specific requirements, remove redundancies, and assign ERP system parameter values. The proposed two-stage solution was validated experimentally using real implementation data and publicly available ERP documentation. The results confirm that selected LLMs, particularly ChatGPT 4o, Copilot, and Qwen 2.5-Max, can perform these tasks with high accuracy and significantly reduce processing time. These findings support the conclusion that integrating LLMs into ERP implementation processes can streamline early-stage analysis, lower vendor workload, and mitigate the risks associated with shifting client requirements..