| KEYWORD |
Area Engineering
Exploring the Role of Large Language Models in Enhancing Decision-Making for Logistics and Supply Chain Management
keywords ARTIFICIAL INTELLIGENCE, DIGITAL TWIN, LARGE LANGUAGE MODELS, LOGISTICS, MACHINE LEARNING, SUPPLY CHAIN
Reference persons GIOVANNI ZENEZINI
Research Groups www.reslog.polito.it
Thesis type EXPLORATORY RESEARCH, LITERATURE REVIEW, RESEARCH
Description LLMs, such as GPT, BERT, and others, offer powerful tools for natural language understanding, generation, and decision support across a range of applications. Their strengths lie in their versatility, fluency, and wide-ranging knowledge, making them valuable for tasks like automated customer service, content creation, and decision support systems. However, their limitations—including hallucination, bias, resource intensity, and a lack of real comprehension—present challenges for sensitive or critical applications, such as in supply chain management or high-stakes decision-making, where accuracy, reliability, and fairness are essential.
The thesis will analyze how LLMs, as conversational agents, can improve decision-making in logistics and supply chain management context, also in combination with other AI techniques (e.g., Machine Learning, Neural Networks, Generative algorithms etc.) and enabling technologies such as Digital Twins.
The thesis will make use of a set of tools including a literature review of scientific and grey literature, comparative analysis of existing LLMs, surveys and interviews to decision-makers and the development of a conceptual framework.
Required skills Basic knowledge of artificial intelligence tools
Knowledge of the reference field (Logistics and Supply Chain Management)
Ability to process documentary data
Deadline 08/10/2025
PROPONI LA TUA CANDIDATURA