Towards Next-Generation Urban Decision Support Systems through AI-Powered Construction of Scientific Ontology Using Large Language Models—A Case in Optimizing Intermodal Freight Transportation

Highlights: What are the main findings? We have developed an integrated and automated methodology that leverages a pre-trained Large Language Model (LLM) to generate scenario-based ontologies and knowledge graphs from research articles and technical manuals. Our methodology utilizes the ChatGPT API as the primary reasoning engine, supplemented by Natural Language Processing modules and carefully engineered prompts. This combination enables an automated tool capable of generating ontologies independently. The ontologies generated through our AI-powered method are interoperable and can significantly facilitate the design of data models and software architecture, particularly in the development of urban decision support systems. What is the implication of the main finding? We compared ontologies generated by our LLM with those created by human experts through CQ-based qualitative evaluation, assessing the reliability and feasibility of our approach. The methodology has been successfully applied to intermodal freight data and simulations. This has allowed us to generate a scenario-based ontology and knowledge graph that enhances data discovery, integration, and management, thereby supporting network optimization and multiple criteria decision analysis. Our methodology is both generalizable and adaptive, enabling the automation of ontology generation to support the development of urban and environmental decision support systems across various disciplines. The incorporation of Artificial Intelligence (AI) models into various optimization systems is on the rise. However, addressing complex urban and environmental management challenges often demands deep expertise in domain science and informatics. This expertise is essential for deriving data and simulation-driven insights that support informed decision-making. In this context, we investigate the potential of leveraging the pre-trained Large Language Models (LLMs) to create knowledge representations for supporting operations research. By adopting ChatGPT-4 API as the reasoning core, we outline an applied workflow that encompasses natural language processing, Methontology-based prompt tuning, and Generative Pre-trained Transformer (GPT), to automate the construction of scenario-based ontologies using existing research articles and technical manuals of urban datasets and simulations. From these ontologies, knowledge graphs can be derived using widely adopted formats and protocols, guiding various tasks towards data-informed decision support. The performance of our methodology is evaluated through a comparative analysis that contrasts our AI-generated ontology with the widely recognized pizza ontology, commonly used in tutorials for popular ontology software. We conclude with a real-world case study on optimizing the complex system of multi-modal freight transportation. Our approach advances urban decision support systems by enhancing data and metadata modeling, improving data integration and simulation coupling, and guiding the development of decision support strategies and essential software components.