This paper attempts to introduce the outline, basic concept and methodology of three-dimensional (3D) geologic modeling in the petroleum industry, especially focusing on 3D reservoir modeling.Since obtaining precise subsurface geologic model is crucial for successful petroleum exploration and development in the petroleum industry, multi-scale modeling methods, including geochemical basin burial history modeling, geologic structure modeling, stratigraphic modeling and reservoir modeling, are requisitely applied to simulate complex geologic features and processes. 3D reservoir modeling aims for reservoir-scale rock body modeling, in which multidisciplinary approaches based on sedimentology, sequence stratigraphy, seismic geomorphology, exploration geophysics and geostatistics are applied. The methodology of 3D reservoir modeling comprises two major steps: geologic framework modeling and geostatistical property modeling. The first step, geologic framework modeling, commonly utilizes the sequence stratigraphic concept to construct a stratigraphic framework, and uses recently advanced techniques of seismic geomorphology on 3D seismic survey data to provide useful information on paleo-depositional topography and reservoir distributions. The second step, geostatistical property modeling, aims to quantify the geologic uncertainty, and simulates a 3D quantitative property model within the prepared geologic framework model. The geostatistical property modeling starts with the integration of data on seismic attributes and well-log petrophysics, and subsequently executes geologically constrained geostatistical stochastic simulations to obtain 3D property distributions with conditioning at the data points. Recent efforts in geostatistical property modeling focus on integrating geology and sedimentology with the geostatistical stochastic simulation methods, and on developing various new methodologies such as multi-point, surface-based and depositional process model-based geostatistical modeling, to obtain realistic and precise modeling results. In addition to the development of new methodologies, the selection of appropriate modeling workflows and procedures is crucial for successful modeling, in consideration of the purpose of the modeling, depositional system of the target, and data condition in density, quality and distribution.