In order to solve the optimal design problem of metropolitan bus transit network, a network optimization model was proposed, which had considered the urban traffic congestion. Firstly it analyzed the malpractices of the shortest path strategy in large cities and reasons of traffic jam. It studied complex network theory and proposed the bypass strategy based on edge betweenness. Through extending the edge betweenness, it defined effectivity edge betweenness and effectivity path. Then it proposed metropolis transit network optimization model based on edge betweenness and designed its algorithm. Take the street network of Changchun City for example to solve. The results show that with the changes of β, the average trip distance first decreases and then increases. When β=0, it is shortest. The average travel speed first is basically stable and then increases sharply. When β=1 the residents average travel time is shortest and the network efficiency is highest. At this time optimization model can rapid decrease the losses of network efficiency caused by traffic congestion.