Analysis of Carbon Emission Reduction Effects by Future Mobility Adaptation Scenarios Using a Large Language Model

Abstract

Mobility refers to mobile possibility and various mobility activities, including various means and services. The development of mobility affects the improvement of accessibility within the city and impacts not only the spatial structure but also the environmental pollution and social equity of the city. Self-driving cars and urban air mobility have recently emerged, and various influences on the environment and society have become issues. The emergence of GPT, a conversational artificial intelligence chatbot developed by OpenAI, a leading AI research foundation in the United States, in November 2022 has sparked a surge of activities integrating artificial intelligence across various fields. This study therefore aims to build an optimal future mobility scenario (model) by deriving and designing various mobility introduction scenarios using prompt engineering based on Generative AI. Using Generative AI, it is possible to create various scenarios at a low cost, in a short amount of time, and to envision scenarios and analyze effects based on various conditions through user-based prompts. To this end, this study creates a database drawing on previous literature, factors affecting mobility change and use, evaluation indicators, policies, and businesses to use for scenario development and evaluation. The findings will subsequently be applied to various Large Language Models based prompt tools such as GPT 4.0 and Llama-2 to configure various scenarios and to compare them to each other to identify the most optimized scenario. Finally, we will propose a framework that develops a set of scenarios based on the user’s prompts and predicts future effects in order to analyze the carbon emission reduction effects. The research results will be used as basic data for future city policies and plans aimed at carbon neutrality and will contribute to sustainable urban development.