Improving Solutions of Shortest Path Problem with MGHS Algorithm
Lei Wang, Xin Wang, Yufeng Yi · 2016
The modified global harmony search algorithm (MGHS) is designed to solve the shortest path problem. The dynamic gene mutation rate is defined to add into harmony search algorithm to avoid producing local optimal solutions. The path is constructed based on priority value of corresponding node in harmony vector with dynamic priority value encoding rule, and the shortest path is obtained by updating harmony memory. Finally, experiments are simulated on 20 to 100 nodes of network topology. Three performance indicators are designed to analyzing performance of IGHS, HS and PSO. Experimental results show the proposed algorithm is superior to HS and PSO for the shortest path problem.