A comprehensive multi-objective approach of service selection for service processes with twofold restrictions

To build complex business processes with QoS constraints by services with inter-service correlations is a challenge. Meeting this challenge involves the QoS-aware service selection problem for service processes with twofold restrictions (QoS constraints and inter-service correlations). Recently, QoS-aware service selection has been addressed from a multi-objective perspective. However, existing research fails to comprehensively consider both QoS constraints and inter-service correlations that widely exist in a real business world. Besides, none of them comprehensively considers both service selection at build time and run time, which may lead to high economic compensation caused by breach penalties at run time. In this paper, we present a comprehensive multi-objective optimization model for service selection with twofold restrictions. Furthermore, an improved multi-objective evolutionary algorithm framework for this model (MOEAQI) is proposed, which can provide a systematic solution for service selection at build time and run time respectively. In this framework, a constrained-domination principle is designed for the specific objectives and constraints in different phases, and several improved strategies are proposed to increase the searching effectiveness and efficiency. Finally, three sets of experiments with different problem sizes are constructed, and our approach is tested in comparison to the state-of-the-art algorithms for service selection problems including GDE3, E3MOGA and DASC. The performance of each algorithm is evaluated in the aspects of set coverage, spacing and computation time. Experimental results demonstrate that our approach is more effective than existing methods for service selection with twofold restrictions.