Development of the Upper Permian high viscosity oil deposit is expected to involve thermal methods of enhanced oil recovery, in particular, cyclic steam treatment of wells. The simulator used for the deposit modeling is to be capable of computing the change in the rock fluid thermal properties, and of solving the heat and mass transfer equations. The modeling of the cyclic steam treatments used CMG STARS, a numerical simulator with a wide range of applications including modeling of thermal processes. The cyclic steam treatment includes three basic stages, such as the steam injection period, the soak period, and the production period. The number of cycles has been selected by computation on the basis of an optimum well operation. To determine the optimum quantity, computations with 7, 5 and 3 cycles were performed. Another predictive computation was performed for a well operation without cyclic steam treatments. The computations helped to determine the operation scenario with the highest cumulative oil production. During the assessment of the cyclic steam treatment efficiency by using hydrodynamic modeling, we obtained dependencies on a number of geological and physical factors, such as steam dryness fraction, formation thickness, steam injection rate, soaking time. The numerical experiments resulted in conclusions and recommendations concerning the case-by-case approach to selecting optimum parameters of the cyclic steam treatments for each individual well, taking into account the structure and specific features. Factor analysis was used to select the optimum parameters for the cyclic steam treatment of the wells drilled in the Upper Permian deposit of the Usinskoye field. For comparison, three predictive scenarios of the well operation have been computed. Cost-performance indicators of the well operation scenarios in the Upper Permian deposit of the Usinskoye field were evaluated, assuming that the wells will be operated in conditions of the natural recovery drive and multiple cyclic steam treatments, in the baseline and recommended scenarios.