Contemporary knowledge about civilizations is fragmented across numerous specialized disciplines — history, archaeology, sociology, economics, political science, astrobiology, and others. There is no unified theory of the civilizational process, no commonly accepted system of parameters for comparing civilizations, and no reliable methods for forecasting their long-term dynamics. Civilizationology is proposed as a meta-discipline designed to synthesize the data from these sciences into a unified explanatory model. The object of study is civilization in the broadest sense: a large-scale form of social organization characterized by large population, developed division of labor, writing (or other systems of external symbolic information fixation), cities, and suprabiological institutions. The subject of study is the universal patterns of genesis, structure, dynamics, and interaction of civilizations. To transition from qualitative description to quantitative analysis, primary parameters are proposed: population size, per capita energy consumption, urbanization level, literacy level, Technological Complexity Index, degree of social stratification, and environmental transformation coefficient. Methods include historical and archaeological reconstructions, comparative analysis, mathematical modeling (agent-based models, network theory), and SETI methods. Fundamental tasks include developing a general theory of civilizational dynamics and a taxonomy of civilizations. Applied tasks on the earthly horizon include civilizational audit tools, forecasting global risks, and long-term development scenarios. On the cosmic horizon, the task is to develop predictive models of technosignatures for the search for extraterrestrial civilizations. Civilizationology does not replace existing sciences but serves as a meta-disciplinary framework for their synthesis. Its development is an urgent task for 21st-century science.