Abstract:With the construction of high-speed railway network in China, the train operation environment is more complex and changeable, which puts forward higher requirements for the daily traffic dispatching. Focusing on delays caused by sudden events such as strong wind, rain, snow and equipment failure, this paper investigates how the train running can be intelligently and efficiently restored according to the operation diagram by adjusting train running orders and arrival/departure time, yet without changing the train running path. This paper innovatively models the problem of train travel time adjustment with multiple constraints, such as arrival and departure interval and overtaking, as an optimal path search problem on the three-dimensional, an improved ant colony algorithm is proposed to realize the optimization of high-speed train dispatching, and a dynamic adjustment method of information heuristic factors and expectation heuristic factors for high-speed train dispatching is proposed, in order to improve the convergence speed and maintain the solution quality. The simulation results show that the “time$ = $space” conversion model and the weight adaptive adjustment method proposed in this paper can effectively improve the performance of ant colony search to solve the high-speed train scheduling problem, and can realize the optimization of high-speed train operation scheduling.