For the bottleneck relief of urban rail transit station infrastructure is lack of quantitative system analysis and its cost is uncertain, a quantitative analysis model for station bottleneck was put forward, and based on which a new optimization strategy was advanced. First of all, passenger train flow diagram was established, and based on the series-parallel hybrid queuing network system optimization model was constructed; secondly, on the basis of the existing optimization strategy, a new control optimization strategy was put forward: change sequence, namely exchange machine physical sequence of security check device and automatic ticket gate; lastly, in Shanghai Xinzhuang subway station according to the two kinds of optimization strategy, specific optimization schemes were advanced and then simulated. Three optimization schemes effectively reduce the passenger queuing time, but the total cost difference is very big. As for a certain arrival rate, compared with no optimization scheme, optimization scheme one with adding one security device reduced the waiting time by 92.5%, increased the total cost by 3.2%; optimization scheme two with exchanging the sequence of security check device and brake machine, reduced the waiting time by 80.3%, decreased the total cost by 50.4%; and optimization scheme three, the composition of scheme one and two, almost completely eliminated the queue waiting time, but increased the total cost by 29.6%. The result analyses show that the proposed model can well simulate the bottleneck relief cost, and the new strategy is superior to the traditional strategy in reducing the total cost.