ChatGPT und Co. - Analyse der Einsatzpotentiale von Large Language Modellen in der Informationsverarbeitung im Supply Chain Management

  • Fabian Hofer

Student thesis: Master's Thesis

Abstract

In an increasingly digitized and globalized economy, supply chain management faces progressively complex challenges. In light of global crises, heightened competitive pressure, and demographic shifts, the question arises as to how new technologies, particularly large language models, can be leveraged to optimize information processing within supply chains, thereby automating tasks and conserving resources. This study is motivated by the desire to explore the potential of this technology to improve information flow and enhance the efficiency of supply chains, ultimately contributing to the advancement and digitalization of supply chain management. Methodologically, the theoretical foundations of large language models and their potential applications in supply chain management are analyzed. The technology is then examined in the context of information flow within supply chains to highlight the enhanced possibilities for efficient information processing when implementing this technology. A central part of this work is the empirical investigation through an experiment that compares the performance of large language models with traditional methods of information processing. By contrasting these approaches, the potential of this technology can be assessed and discussed. The fundamental methodology includes an extensive literature review and a controlled experiment in which both human and AI-based groups solve tasks related to supplier offer processing. The results demonstrate that large language models have the potential to significantly enhance efficiency in information processing. In particular, tasks could be completed more quickly and with comparable or higher accuracy than traditional manual methods. This suggests that LLMs could play a crucial role in supply chain management in the future, especially in areas where rapid and precise information processing is essential. The findings are particularly noteworthy in the context of the various evaluations and analyses, where it was observed that even with a stronger emphasis on accuracy, large language models perform exceptionally well. Furthermore, the results have implications for the theoretical framework of supply chain management, as this technology should be considered in the future design of supply chains as an expansion of information technology capabilities. They also have practical implications, as the implementation and use of large language models in various application scenarios appear to be beneficial. This study makes a significant contribution to understanding the application of artificial intelligence in supply chain management and demonstrates that the use of large language models can lead to an optimization of supply chain processes. Further research in other areas of application, with a larger sample size or by including various large language models, could provide additional insights into the application and potential of this technology.
Date of Award2024
Original languageGerman (Austria)
SupervisorPatrick Brandtner (Supervisor)

Cite this

'