<p>The metro plays an important role in urban public transport, while frequent unplanned service disruptions significantly affect the efficiency of the metro system and passenger satisfaction. Passenger flow guidance, as a flexible emergency management measure, can adjust the metro passenger flow distribution through influencing passengers’ travel behavior. With the development of information technology, it has become possible to push personalized guidance information for passengers. However, the mechanism by which personalized guidance information influences passengers’ behavior during metro service disruptions remains unclear. Therefore, we rebuild a detailed decision-making process of passengers under metro service disruptions and design personalized guidance information strategy. Further, considering attitudinal preferences, we develop an integrated choice and latent variable (ICLV) model to reveal the effect of guidance information on travel choice behavior based on the survey data collected in Beijing, China. The results show that the provision of guidance information significantly affects passengers’ travel choice. Compared to low information intensity that involves pushing a single message to passengers, medium information intensity that involves repeating multiple messages has a positive effect on passenger compliance with recommended travel plans. Additionally, passengers with stronger team consciousness are more likely to choose the travel plans which is recommended to alleviate overall metro congestion, and passengers with higher satisfaction of information benefits are more inclined to agree with the role of guidance information and to choose the recommended plans beneficial to individuals. The findings provide valuable insights into the design of personalized guidance information under unplanned metro service disruptions, thereby helping to mitigate their negative impacts.</p>