The technical and commercial success of SAGD projects over the past decade has opened the door to the development of a large number of bitumen reservoirs in Canada, previously thought uneconomical to produce. Some of these reservoirs have overlying gas caps and/or water zones. Some studies have suggested that gas-cap production might "sterilize" the underlying bitumen. Many such studies however, assumed rather thick continuous pays with high permeability, and considered an infinite gas-cap. In this work, a simulation study was conducted to examine the feasibility of bitumen production from a certain project area in Alberta, using the SAGD process, and to study the effect of production from the gas cap. A decision needed to be made as to whether gas production should be delayed until after bitumen production. The large well spacing did not allow a detailed description of the connectivity of the shale layers. The uncertainty was compounded by the geological setting of the study area, a system of channel sands cut through the original marine sand and shale deposits. Since the actual shale connectivity and thickness was unknown, a methodology was developed to incorporate different geological descriptions using the available core and log data. Five reservoir models were developed. Bitumen recovery, average oil production rate, and cumulative steam-oil ratio (SOR) obtained from thermal simulation were the three main-parameters used for evaluation of the attractiveness of botumin recovery operations. These numbers were compared with some of the corresponding values reported and/or forecast for economically feasible operations such as the UTF and Christina Lake projects. The effect of pressure reduction (caused by gas-cap production) on production rate and SOR was also investigated. The results indicated that for conditions considered in this study the effect of gas production on bitumen recovery was minor, and appeared as a small deceleration of the recovery and a small increase in SOR.