The report presents summaries of technology development for discrete feature modeling in support of the improved oil recovery (lOR) for heterogeneous reservoirs. In addition, the report describes the demonstration of these technologies at project study sites. In the discrete feature network (DFN) approach, conductive and flow barrier structures are modeled explicitly, facilitating a more realistic analysis of flow. The DFN approach models these features at all scales simultaneously, from the individual fractures intersecting a well to the major faults which define the reservoir. These discrete features are extrapolated from geological and geophysical measurements using a combination of geological and geostatistical methods. During this project, Golder Associates'team improved the data analysis and geological modeling capabilities of the DFN approach, and developed practical applications of that approach for lOR in heterogeneous reservoirs. Major accomplishments of this research project are summarized in Table 1-1. These include development of approaches for gel treatment, strategic completion, and water control in heterogeneous reservoirs. The discrete fracture network technologies developed within this project are directly applicable to a large percentage of the secondary and tertiary oil recovery projects currently underway in the United States, and can be of significant value to the design of future oil recovery projects. This project achieved significant technological advances in the development of the discrete fracture network modeling approach for heterogeneous oil reservoirs.