Workflow technology is regarded as the automation of an execution process in which the information or tasks are passed from one service to another, according to the predefined execution sequence. Recently, workflow technology has been successfully used in wireless sensor networks (WSNs) for service composition. Although workflows can dynamically change according to current context, there is still limited work to facilitate the atomic services reuse.In this paper, we propose a reasoning-based context-aware workflow management(Recow ) approach, in which a rule-based reasoning module is responsible for extract semantic information so that the lower sensor data will have a loose couple connection with the upper logic process. By using the semantic information as service I/O, we reconstruct atomic services, then they can be reused in workflow construction. Usually more than one rule will be matched and they can not be executed simultaneously in the rule matching process, so a conflict resolution algorithm is further proposed based on context-aware priority. Finally, two case studies demonstrate that our approach can effectively facilitate resource reuse and also indicate the performance of the Recow approach as well as the precision of the conflict resolution algorithm.