The invention discloses an urban railway traffic random passenger flow loading method which comprises the following steps: 1.1, initializing, dividing one day into a plurality of
time windows averagely, loading basic data, at the same time, setting the total number of the
time windows as n, setting and marking an initial value of a variable t of the
time windows as 1, and setting an initial value CN1 of
road congestion charge as 0; 1.2, reading an origin and destination (OD)
distribution volume of a tth time window, based on a basically effective path collection, calculating time of a first regular
bus and a last regular
bus of each path, judging the relationship of the time and the tth time window, and generating a dynamic effective path collection; 1.3, based on the dynamic effective path collection and an urban railway traffic passenger going-out path selection model, confirming a selected portion of each effective path between the OD, and obtaining a path flow; 1.4, performing statistics on a passenger flow of each section of a road, and updating the
road congestion charge CNt+1; and 1.5, stopping the judgment, if the t is less than n, t=t+1 and continuing to iterate from the step 1.2, or, stopping the method.