There are many existing routing strategies in complex networks, but there is no uniform standard to measure whether the strategies achieve optimal transmission effect. A pervasive optimized algorithm is proposed. The key factor restricting transmission capacity is maximum betweenness centrality and minimizing it becomes the uniform standard. In order to make betweenness centrality more evenly distributed and balance the traffic load of each node, we use punishment selection method to avoid the nodes with larger betweenness centrality. The simulation results show that the new algorithm could reduce maximum betweenness centrality of existing strategies and improve the network transmittability greatly.