Abstarct: In the petroleum exploration and production industry, characterizing the reservoir quality, identifying the main rock types and predicting their lateral variations is a main challenge. Beside subsurface structural information, 3D seismic data contains valuable lithological information. Utilizing seismic attributes to subsurface analysis can reveal important features of regional geology to detailed reservoir properties in the form of seismic facies classification. Since it is difficult for interpreters to extract and combine all the available information from seismic attributes, classification methods have been developed to evaluate the explorative targets and to improve reservoir characterization within the field development projects. Therefore the objective of the facies classification process is to describe subtle characteristics within the seismic data and relate this to lithology and ultimately rock properties to help identify potential hydrocarbon accumulations. Identifying lithological facies and their lateral distribution in reservoir level through determination of various seismic facies and their spatial distribution by integrating well data, geological information and 3D seismic data and using multi-attribute analysis baxsed on two unsupervised, competitive learning algorithm, and supervised, error back propagation, neural networks classification approach is the main object of this study. The described procedure has done for one of the Iranian exploration fields in Persian Gulf, as at first unsupervised analysis has been done to get a general idea about the distribution of seismic facies and then the results of unsupervised classification are being used to define training data for supervised classification. Volume baxsed and VRS attributes have been extracted from reservoir interval between top and baxse horizons. Statistical analysis of different seismic attributes in conjunction with geological significance of seismic attributes were the main criteria for selecting the appropriate set of seismic attributes for the classification. The results of this thesis have been increase effectively our understanding about the distribution of different seismic facies and reservoir heterogeneity in the interval under study. The results indicate the presence of Shaly and Sandy facies in reservoir zone. Especially in the location of four wells which reach to the Oil, Sandy facies with interlaxyers of shale are dominant and checking these zones for the presence of Oil reservoir is important.