An Encoding Algorithm Based on the Shortest Path Problem

Abstract

The transportation network of the city is dynamic and stochastic, The problem of dynamic stochastic shortest path is NP-hard. the optimal problem of path is widely used in the fields of transportation, communication and computer network. An improved self adaptive genetic algorithm is proposed by encoding the chromosomal mode.The paper investigates the shortest path problem based on the genetic algorithm principle, and improved genetic algorithm by adjusting the encoding parameters. Mny experiments indicate that the improved genetic algorithm could adapt to new transportation rapidly in global optimization than A* algorithm and Dijkstra algorithm and obtain the better solutions in the shortest path problem in the fields of transportation and computer network.