To promote the deep integration of artificial intelligence technology with transportation planning, operation management, and urban governance, and to effectively connect research achievements in transportation large models with actual industry needs, an industry matchmaking conference (roundtable forum) on "Transportation Large Models and Decision Support for High-Quality Development" was successfully held at the Sipailou Campus of Southeast University on July 18, organized by the Jiangsu Province Collaborative Innovation Center of Modern Urban Traffic Technologies. Experts, scholars, and industry representatives from universities, research institutions, industry management departments, and planning and design institutes attended the meeting. Wang Wei, Director of the Jiangsu Province Collaborative Innovation Center of Modern Urban Transportation Technologies, delivered the opening address and gave a keynote presentation.
Currently, AI, big data, and AI agent technologies are accelerating their entry into fields such as transportation planning, signal control, traffic simulation, travel services, and urban governance. Faced with the complex characteristics of transportation systems characterized by multi-stakeholder participation, multi-scenario coupling, and multi-objective coordination, traditional planning and management approaches relying on experience and single models urgently require transformation and upgrading. Big models in transportation, by integrating multi-source data, professional knowledge, analytical models, and simulation tools, provide new technical pathways for traffic situation awareness, demand forecasting, plan generation, effectiveness evaluation, and decision support.
At the meeting, representatives from Harbin Institute of Technology (Shenzhen), the Road Traffic Safety Research Center of the Ministry of Public Security, Zhuhai Institute of Urban Planning and Design, Guangzhou Transport Planning Research Institute Co., Ltd., Institute of Automation of the Chinese Academy of Sciences, Hangzhou Zhituo Yunkong Technology Co., Ltd., and other institutions engaged in in-depth exchanges regarding the theoretical methods, technical architecture, business scenarios, and engineering applications of transportation large models. Drawing on practices in transportation science research, urban traffic signal control, planning data analysis, and intelligent decision-making platform development, participants explored the practical pathways for transforming large models and agent technologies from general capabilities to transportation-specific professional capabilities.
Participants noted that the application of transportation large models must move beyond conceptual demonstrations, report generation, and single-function showcases, and instead address real-world operational needs such as signal control, demand analysis, planning evaluation, and emergency incident response. Industry organizations should further open up their actual demands and testing scenarios, while research teams need to enhance the standardization, modularization, and engineering adaptability of their technical outputs, thereby promoting joint collaborative problem-solving between both the supply and demand sides.
This conference further strengthened exchanges between research teams and industry organizations. Moving forward, the Jiangsu Province Collaborative Innovation Center of Modern Urban Transportation Technologies will continue to serve as a platform bridge, focusing on industry demand analysis, technology transfer, opening of application scenarios, joint R&D efforts, and demonstration project development. It will promote the establishment of an innovation mechanism featuring collaborative participation among demand-side, R&D, and application stakeholders, accelerate the validation and transformation of transportation large model technologies in real-world business scenarios, and provide scientific and technological support for the smart transportation industry and the development of new quality productive forces.
