Services Derivation From Business Process: A PSO-based Multi-Objective Approach

Software module clustering is generally a difficult and challenging problem in software engineering. In the same way, service identification plays a critical role in service engineering. Existing Service identification approaches are often prescriptive and based on the architect's experience thus could result in non-optimal designs which results in complicated dependencies between services. In this paper we proposes top down approach to identify automatically services from business process by using several design metrics. Service boundaries are identified from business processes by automated search, guided by a multi-objective fitness function and using a clustering combinatorial particle swarm optimization algorithm. In order to evaluate the effectiveness of the proposed approach, a set of experiments were performed. The experimentation results of this empirical study denotes that our approach achieves better results in term of performance and convergence speed.