Change-point detection in the recurrent-event context has been an interesting topic in many fields such as medical studies and travel safety analysis. When subgroups exist, clustering can be incorporated into the change-point detection. This paper develops a Recurrent-K-means algorithm to detect the change-points and cluster the objects in the recurrent-event context. The paper also proposes a heuristic searching method to determine the number of underlying clusters. The objects are assumed to fall in several clusters while the objects in the same cluster share identical change-points. The event counts for one object are assumed to be a non-homogeneous Poisson process with piecewise constant intensity functions. The methodology estimates the change-points as well as the intensity rates before and after the change-point by cluster. It uses the clustering scheme motivated by K-means algorithm, but advances the procedure to be model based. The simulation study shows that the methodology performs well in parameter estimation and in determining the number of clusters under different scenarios. The methodology is applied to the coal mining disaster data in UK as an illustration.