Location models play an important role in location-based services (LBSs), because LBSs require a well-defined representation of location knowledge to support location browsing, navigation and query processing. Current location models can be divided into two categories: symbolic and geometric. Symbolic models try to represent the semantic relationships between entities, and geometric models are based on geometric coordinates and Euclidean distance.In this paper, we propose a lattice-based semantic location model (LSLM) for the indoor environment. LSLM is based on the exit-location model and the theory of “formal concept analysis.” The model can provide an explicit representation of the basic relationships between two entities such as containment and overlap. The nearest neighbor relationship on the concept lattice is used to define the optimal distance between two entities. Furthermore, the dual (location/exit) property of the model can cater for different navigation needs. We provide examples to show the effectiveness of our model.