Summary: Granular computing (GrC) is a new method to simulate human’s thinking in the field of computational intelligence research. GrC includes all theories, methods and technologies related to granularity, and it is an effective tool to solve complex problems. Granular concept is an important notion which is defined by combining granular computing and formal concept analysis. In order to update the granular concepts in a dynamic formal context, this paper introduces the notion of an object-oriented granular concept, analyzes how the extent and intent of the object-oriented granular concept are changed, and puts forward approaches to update the object-oriented granular concepts when objects and attributes are gradually removed from a formal context. Finally, some numerical experiments are conducted to show the effectiveness of the proposed algorithms.