This paper discusses the concepts and methodologies of Geographical Simulation and Optimization System (GeoSOS). The objective is to provide an efficient spatial exploratory tool for geographic process analyses. The basis of GeoSOS includes complex system theories, GIS, computer sciences and geography. The proposed GeoSOS includes three components, Cellular Automata (CA), Multi-Agent Systems (MAS), and Swarm Intelligence (SI). The establishment of GeoSOS can provide useful toolboxes for solving many simulation and optimization problems that can be difficult to solve by using existing functionality of GIS.