SERVICE COMPOSITION BASED ON IMPROVED GENETIC ALGORITHM AND ANALYTICAL HIERARCHY PROCESS

摘要

During the past few years, a service-oriented architecture paradigm was utilized by more and more business entities and organizations, and numerous services were deployed on the Web environment separately. Compositing these decentralized services based on their quality information is a time-consuming optimization problem because of their dynamic nature. To address the problem in a timely manner, a method using the improved genetic algorithm and AHP (Analytical Hierarchy Process) is proposed in this paper. The AHP algorithm is used to select local optimal services, and the mechanism of similar chromosome exclusion is introduced to expand the diversity of population as the "premature" phenomenon may occur during the process of evolution. The strategy of elite individual reservation is adopted to accelerate the convergence of genetic algorithm. The simulation results demonstrate that the improved approach in our work has good convergence and efficiency, and the service composition can be realized more efficient than before.