An Efficient Multiple Sources Single Destination (MSSD) Heuristic Algorithm Using Nodes Exclusions

Abstract

The problem of identifying the best paths between given set of nodes and a given single-destination in a graph of vertices is commonly referred to as network multiple sources single-destination problems. In real life researchers always find themselves in a critical situation that researchers seek the nearest set of related points such as the urgent need for fire stations. This study describes, the problem and proposes an algorithm for finding the shortest paths between the set of sources <si> and a single-destination <t> given that <si> and <t> ∈ weighted graph G(V, E, w) with vertex set V and arc set E associated with non-negative real valued weight. An efficient algorithm is developed based on different graph representations. The proposed heuristic determines a candidate subgraph G' and excludes all nodes that do not lead to destination. The proposed algorithm improves partially the performance of improved traditional shortest path algorithms, i.e., Dijkstra's algorithm. This is shown obviously by applying the algorithm on set of random graphs.