Time-cost optimization problem in service-workflows is a widely existing problem hard to be solved in practical systems. In this paper, an Niche Technique multi-objective PSO method is proposed. Generating initial optimal solutions with single-objective. External Population and Niche Technique and Meshing hybrid method is introduced to obtain an evenly distributed Pareto set. Experimental results prove that the proposed algorithm is efficient and effective. For various characteristic instances, more evenly distributed Pareto Sets are obtained with high quality.

Scheduling the workflow in a dynamic environment is an important issues for actual application because of service resources changed dynamically . In this paper, a cost-compensation-based dynamic scheduling algorithm (CCDSA) is constructed for the dynamic service-workflows. Related rules are defined, local re-scheduling workflow is performed by judging the influence range of abnormal nodes. Comparing CCDSA with HPSO, experimental results show that CCDSA obtains satisfied solutions, and has good adaptability and stability.