The uncertainty of urban road traffic network leads to the fact that the traveler's travel time is a random variable subject to a certain distribution. The uncertainty of travel time is an important factor influencing traveler's choice behavior. In addition, due to the mutual influence between vehicles, insufficient supply of parking spaces and other factors, the parking space seeking time of vehicles also presents uncertainty. According to the analysis results of the questionnaire survey, 60% of respondents believe that the uncertainty of parking space seeking time will affect whether they choose to drive. In this paper, under the condition of network degradation, the travel time of each path in the network is uncertain. In addition, it is assumed that the parking space seeking time of vehicles in the parking lot is a random variable subject to normal distribution. The transportation mode choice behaviors are investigated and analyzed considering the uncertainty of travel time and parking space seeking time in the multi-mode urban transportation network. Based on travel time budget (TTB), a nested logit (NL) model for multi-mode and multi-class user is proposed. A nested method of successive average algorithm was designed to solve the model. Numerical results indicate that the confidence level produces a great influence on decision cost and mode choice behavior of travelers. Travelers with higher expectation of punctuality have larger travel time budget and are more likely to choose subway, while travelers with lower expectation of punctuality are more likely to choose to travel by car. Among the travelers who drive, those who expect to arrive on time with low probability will choose to drive in the whole journey. Accordingly, those who expect to arrive on time with high probability are more likely to choose park-and-ride. key words: urban transportation; travel time budget; nested logit; nested method of successive average

CICTP