Abstract
Artificial Intelligence (AI) is expected to change the project management methodology, and the current rise of Large Language Models (LLM) is an important trend. Project management is still largely conducted by humans who base their discussions and decisions on natural language and text. Therefore, it looks promising to use LLMs for decision support in project management, given the current results of text processing by LLMs. Especially, ChatGPT is widely tried out by scientists and practitioners since it is free of charge and easy to use. This paper investigates the quality of the recommendations given by GPT when it comes to staffing projects. Building on our previously conducted research on data analytics in project management, we study the contribution of LLMs to assigning people to projects based on textual descriptions of projects and staff competence profiles. Having the right people assigned to the right projects is an important goal in project management. We provide a short literature review on using AI in decision support for project staffing, and 2 experiments based on real project and competence data of a German engineering company. We use ChatGPT and a tailored GPT, compare and discuss the results, and explain limitations.