Objectives: To find residential areas with high incidence rate of tuberculosis in Moscow using spatio-temporal analysis of incidence data. Methods: We analyzed the spatial patterns of residence locations of smear or culture positive patients with pulmonary tuberculosis in Moscow. To identify clusters with high local incidence rates, the neighborhoods of detected cases were studied. We assessed the spatial and temporal stability of clusters. Results: For 19 033 cases diagnosed with smear or culture positive pulmonary tuberculosis among residents of Moscow in 2000–2015 we identified 18 small-scale clusters of increased incidence rate responsible for 3% of all registered cases identified on a territory inhabited by only 1% of the population. Locations of clusters were sufficiently stable in space throughout the whole period. The local incidence rate inside clusters was significantly (3–4 times) higher than the city average during the whole observation period. The presence of clusters was associated with the incidence rate in the surrounding area. Socio-demographic characteristics of patients in clusters were not significantly different from the average characteristics of patients in the city. Conclusions: The detected small-scale clusters of increased incidence may be used to target active case finding for tuberculosis. The causes and mechanisms of cluster formation and stability need further study.