Аннотация:Under conditions of the Krasnoleninsky field (hydrocarbon producing territory in Western Siberia), the pay zones of the Tyumen formation are characterized by a thin-layered structure and a high degree of spatial variability of petrophysical and lithological characteristics. Due to the low elastic contrasts of reservoir rocks relative to the embedding sand-clay sediments, their prediction from seismic data is challenging. Additional complications are associated with the presence of a significant amount of coal material in the geological section. Minimizing the uncertainties of the pay zones prediction can be achieved by integration a priori geological information, seismic and well logging data. Geological information includes knowledge on depositional conditions and evolution of the sedimentary basin that manifests in vertical and lateral facies successions and trends. The paper presents results of geologically driven stochastic inversion of 3D seismic reflection data at one block of the Krasnoleninsky fields. The proposed approach is based on the parameterization of the available geological information and its integration with the results of the petrophysical interpretation of the well logging data within the framework of stochastic inversion.