The shortest path problem is the problem of finding a path with minimum total weight from a source node to each destination node in a network. The existing solution to this fundamental problem searches the shortest paths to all network nodes until it meets the given multiple destination nodes. By granting preference to routes of each destination node, the proposed algorithm meets the destination nodes faster. Compared to the existing solution, the average time improvement of the proposed algorithm is in order O(n), where n is the number of nodes in the given network. The experimental analysis results on real-world datasets and simulated random networks show the superiority of the proposed algorithm to the existing solution. This remarkable improvement makes the proposed algorithm applicable in all related applicatio