The present invention relates to a method for estimating distribution of urban road travel time in considering operation state of taxi, and belongs to the technical field of urban transportation planning and management. The distributions of path travel time are respectively estimated according to different operation states of the taxi. When the distribution of the path travel time is estimated, adjacent road sections in the road network are not independent. In the present invention, the Markov model is added to describe the correlation of the travel time distribution between the adjacent road sections, so as to increase science and accuracy of the estimation result. In the present invention, the weight is set according to the proportion of the number of the vehicles under two different operation states to obtain the final distribution of the path travel time. Driving behaviors of taxi drivers under two operation states of no passenger and passenger service may be different, so a difference between the travel time estimated from direct use of taxi data and a true value inevitably exists. The proposed model is calculated through Shenzhen data, which shows that the calculated travel time distribution function is more accurate after considering the operation states.