Enhancing Service Design: The Potential of Large Language Models for Innovation

Service design is a well-established discipline across many industries, valued for its capacity to improve user experience by employing a methodology and mindset that embrace a wide range of interpretations, making it applicable to any human activity. This thesis examines service design from a practical perspective, focusing on its most widely used tools, which will be experimented in the second core topic of this thesis: Large Language Models. Specifically, there is an attempt of explaining Large Language Models by simplifying the architecture of a Generative Pre-trained Transformer (GPT). This explanation serves as the groundwork for the final section, where customizable GPTs are used to enhance traditional service design tools. The objective of this study is to discover new approaches to leverage this emerging technology within the field of service design, providing tangible solutions for industrial applications in order to save time, produce better results, coordinate the team and provide several insights.