Abstract: Objective: To fulfill the goal of enhancing the satisfaction degree of multiple stakeholders in a project, improving project implementation efficiency, promoting scientific evaluation of URT (urban rail transit) project performance, encouraging public participation in URT projects, and further implementing the concept of ′coconstruction, cogovernance, and sharing′ in the field of infrastructure project governance, it is necessary to establish a performance evaluation indicator system for URT projects based on public demands. Method: The literature screening method is employed to select indicators related to the public from existing literature, achieving the initial indicator selection. The text mining method is applied to extract indicators from online messages expressing public interest demands, optimizing the initially selected indicators. The interview method is used to collect opinions from the general public of all age groups, supplementing indicators and thus establish a preliminary evaluation indicator system. The characteristics of the obtained indicators are used as clustering criteria to perform clustering calculations and analysis separately for the construction phase indicators and operation phase indicators. The evaluation indicator system is optimized based on clustering dendrogram. Public opinions on the importance of indicators are collected through a questionnaire survey, and the entropy weight method is used to process the data and assign weights to the indicators. The influence of the main indicators is analyzed based on their weights. Result & Conclusion: A performance evaluation system for URT projects based on public participation is established, which includes four primary indicators and nineteen secondary indicators.