Prediction of fluid production based on static reservoir models often fails to match the observed field production history. For reliable future performance predictions, it is imperative that the reservoir models adequately reproduce the past performance history of the reservoir and for this, it is necessary to condition the static model with dynamic reservoir data, such as transient pressure, tracer response and multiphase production history. Integration of dynamic data can be particularly effective in identifying preferential flow paths or barriers to flow that can adversely impact sweep efficiency. Identification of the location and distribution of these barriers is critical to the success of the secondary and tertiary recovery efforts. Integration of production data in conventional finite-difference simulators is computationally infeasible for the large number of simulations involved and has never been done in a fine-scale reservoir model consisting of millions of grid blocks. Lawrence Berkeley National Laboratory and Texas A & M University have, for the first time, adopted a very flexible and efficient approach to merge different data types (production, seismic, well log) by developing a very general and computationally efficient 3D streamline simulator for modeling multiphase flow through fine-scale reservoir models. Besides being cost-effective, the streamline based flow simulator is computationally efficient in rapid inversion of dynamic production data by using an analogy between the streamlines and ray tracing in seismology. This allows application of efficient inversion technology from seismic for integration of production data. The algorithm developed has reduced the time required to invert tracer data for 3D reservoir structure in an entire field by about 10-1000 times. The feasibility of the approach developed has been demonstrated in field applications in the carbonate reservoirs at the North Robertson Unit and the Goldsmith San Andres Unit in West Texas. The new reservoir characterization technology will lead to efficient management of a wide variety of oil and gas reservoirs by identifying bypassed oil/gas for targeting infill drilling, extending life of wells by reducing early water breakthrough and reducing water cycling and enhancing sweep efficiency. Enhanced production of even 1/2% of oil and gas from already discovered reserves by this technology will result in additional production of 400 MMBLS of oil and 26 Tcf of natural gas. This production will generate 17 billion dollars of income for the Federal government in the form of taxes and royalties and will create 9,000 new jobs in various sectors.