Genetic Algorithm with Dynamic Stopping Criteria for Global QoS-Driven Web Service Replanning Selection

As a kind of service selection algorithm, Genetic Algorithm (GA) is a good way to select an optimal composite plan. But, the computation time of GA is always so much that its computation cost may be critical during replanning selection phase. It is necessary that the computation cost spent on replanning selection should be shortened to optimize the execution duration of the remaining functions. To remedy this situation, a novel GA is presented. In this GA, a dynamic stopping criterion, an optimal individual selection and replacement policy are proposed to consider the optimal selection result as well as the computation cost of replanning algorithm. By the running of this GA, these policies can ensure users' QoS requirements during replanning selection. The simulation results on web service selection with global QoS constraints have shown that the constraint on the execution duration of composite service can be ensured effectively.