SUMMARY
One crucial parameter related to subsurface formations fluid flowing is the rock permeability. Generally, rock permeability reflects the formation capability to transmit fluid. Its significance reflected through several methods existing utilized to predict it, including rock core measurements, empirical correlation, statistical techniques, and other methods. The best and more exact permeability findings are acquired in the laboratory from core plug cored from a subsurface formation. Unfortunately, these experiments are expensive and tedious in comparison to the electrical and electronic survey techniques as wireline well logging methods, for example, not exclusively. The current study compares and discusses different methods and approaches for predicting permeability via wireline logs data. These approaches include empirical correlations, non-parametric statistical approaches, flow zone indicator FZI approach. In this research, we introduced a comparatively new process to predict permeability by the combination of FZI method and the artificial neural networks method. All these approaches are performed using well logs data to the non-homogenous formation, and findings are placed in comparison with permeability from laboratory experiments, which is regarded to be standard. Several statistical criteria, such as ANOVA test and regression analysis, were used to determine the reliability of calculated permeability results.