This paper proposes an efficient technique for computing shortest path in dynamic graph. Which finds shortest path in a given graph which is static or intended to change its weight frequently. If that graph is static i. e. not changing its weight then SPT is being calculated once and that remains same. If graph is dynamic i. e. changing its weight then this technique finds new SPT with traversing minimum number of nodes or vertices. This technique extends a few state-of-the-art dynamic SPT algorithms to handle multiple edge weight updates, and find the SPT. A function based on the location of current node/ state is used to vary the cost of the goal node and the search is done with minimum the state space and exploring only affected nodes, by using these approaches problem is solved in minimum time. Based on experimental results on sample data set we propose to device an algorithm which efficiently handles different traffic conditions. The performance of this algorithm is measured on the basis of Graph size, number of changed edge (NCE). To evaluate the proposed dynamic algorithm, comparison is done with the well-known static Dijkstra algorithm. Where proposed algorithm's complexity is O(bd) in worst case O(E) in average case and O(1) in best case.

Dynamic Dijkstra Dynamic Graph Graph Directed Graph.