The train stop plan is usually optimized based on the constraint of empirical service level. With provided routes, this paper constructs an optimization model of stop planning based on the optimal passing-stop ratios to minimize the total passenger travel time. The model has convictive and accurate service level constraints, and the randomly generated initial solution has fewer residual passenger flow and shorter average boarding deviation time. Taking Beijing-Shanghai High-speed Railway Line as an example, we adopt the “rooftops” model to assign the passenger flow, generate an initial solution by optimal passing-stop ratios, and optimize it by genetic algorithm. The initial average boarding deviation time is 27.05 min, with 0.01% residual flow, whereas the optimized average boarding deviation time is 22.69 min without residual passenger flow. The results show that the initial solution of the optimization model is more acceptable, and the algorithm can further optimize the passenger travel time at the same time.

CICTP