The Hassi Berkine South (HBNS) field is an undersaturated, low-viscosity, moderate permeability oil field within the center of the Berkine Basin in Algeria (Fig 1). Discovered in January 1995, the HBNS field had first oil production in 1998 from the Trias Argilo-Greseux Inferieur (TAGI) reservoir with a reservoir development plan (RDP) of crestal, miscible gas injection and peripheral waterflood. Full-field water-alternating-gas (WAG), as a secondary recovery process, is currently under investigation for the potential increased oil recovery and higher oil rate. A 14-month miscible water-alternating-gas (MWAG) pilot program successfully provided valuable information for reservoir simulation. This paper describes the application of WAG pilot results in reservoir simulation and the use of the novel miscible interpretation of the Reservoir Saturation Tool (RST) in history matching the observation well data. A fine-grid simulator sector model was used to simulate the WAG pilot program. The pilot observation well data were used to fine tune the reservoir model and establish the criteria for the realistic simulation tool for the full-field WAG evaluation. The technique for correcting observation well-log data to reflect the miscible displacement process was needed to history match the hydrocarbon saturations. The simulations also examined the impact of various reservoir description assumptions on predicted solvent and water behavior at the observation well. Minimum vertical layering requirement was also investigated for realistic WAG simulation. Introduction WAG injection processes have been accepted as a viable Improved Oil Recovery (IOR) technique1. Because of the complexity associated with the WAG process, a pilot program is often implemented before any large-scale application is undertaken2,3,4,5. This paper reviews the field measurements of two MWAG cycles in a WAG pilot within HBNS field and the analysis of these data using reservoir simulation. Proper analysis of miscible processes occurring in the observation well requires understanding of both the phase behavior of hydrocarbons (HC) as well as the interpretation of the fluid distribution. Between 1983 and 1987, an observation well pilot test was conducted in the Prudhoe Bay Flow Station 3 area4. The epithermal neutron log was used in the observation well DS 13–98 for the hydrocarbon saturation measurements. The miscible interpretation was based on static altered oil and altered gas compositions from the equation of state (EOS) simulations. The use of the static altered oil and altered gas compositions for HC saturation interpretation could result in ~50% error in the area that had not been sweeped by the solvent. This interpretation also led to an undesirable mingling of errors inherent in both the measured and simulation results. The standard log-derived HC saturations have been recognized as an immiscible interpretation5 and do not represent the miscible process. These HC saturations are based on static (constant fluid properties in time and space) conditions and need to be corrected to represent the true fluid displacement (miscible) process. Corrections are needed to relate the simulation results to the field measurements and to understand the process occurring within the reservoir. The WAG pilot design and the development of the miscible interpretation of the RST measurements were documented previously5. The geological model, the reservoir sector model, and the miscible interpretation of the RST measurements have been summarized in this paper for completeness.