Production optimization requires a clear understanding of subsurface conditions throughout the field's life. Description of the subsurface through classical 2D techniques is being replaced increasingly by powerful computer technology. Such technology enables construction of fully numeric 3D models of the subsurface, which can be visualized through high-end graphics displays. This 3D modeling technology is having a significant impact on development planning and monitoring activities. The focus here is on the perceived benefits and challenges of using 3D models to optimize production. The paper includes examples of integrated 3D static and dynamic modeling activities carried out on fields undergoing appraisal and development and fields facing abandonment. Case 1 (appraisal) illustrates how the integrated use of 3D static and dynamic modeling during primary development enables petroleum engineers (PE's) to construct alternative reservoir models rapidly and to evaluate the impact of various uncertainties quantitatively. This approach optimizes production by facilitating fast-track developments, design of robust development schemes incorporating optimized well paths, and providing quantified justification for appraisal expenditure. Case 2 (mid field life) illustrates the benefits of using 3D modeling technology to integrate static and dynamic field data, geological knowledge, and scenario analysis concepts to locate bypassed oil and optimize infill well trajectories. History matching offers the geologist valuable feedback on the robustness of the geological model, including aspects such as the size and orientation of reservoir sands. The relative impact of various uncertainties can be quantified and development schemes optimized. Case 3 (end field life) illustrates how. even in very mature fields that are facing abandonment, 3D modeling technology can facilitate data integration with subsequent cross-discipline interpretation. This leads to the successful planning and drilling of horizontal sidetracks to develop bypassed oil. In all cases, the technology used allows sophisticated 3D models of the subsurface to be built through integration of diverse data types and geological knowledge. The resultant models can be upscaled rapidly for dynamic simulation; model optimization through iterative feedback from the reservoir flow simulator can be achieved easily. The facility to plan and ''drill'' wells in the 3D volume enables development scenarios to be optimized both in terms of well trajectory and multidisciplinary teamwork.