Abstract: Most of the existing researches for composite service reselection method are based on the assumption that the tasks involved are independent. However， in practical scenarios， the QoS of some candidate services have correlations with other services， which leads to the reselected results not to be the optimal one in the real executing environment for these existing reselection methods. To solve this problem， a QoS model supporting the correlations between services was given by extending the OWL-S framework. Based on the proposed model， a composite service reselection method considering the correlations among tasks was provided， in which the correlated tasks were regarded as a task unit and the correlated services of each task unit were regarded as its candidate service set. Experimental results showed that compared with the reselection methods considering the tasks independent， the proposed approach could effectively improve the quality of composite service reselected.

Key words: reselection, QoS, correlation, OWL-S, composite service