The electrolyte nature of concrete, while passing electrical current, causes a close correlation between electrical conductivity and the value of porosity, diffusivity and degree of hydration in concrete. Also electrical conductivity of concrete is widely used to assess the corrosion potential in reinforced concrete. Thus electrical conductivity of concrete could be considered to determine a wide spread of mechanical and durability properties and could be beneficially used to monitor the concrete in product and service steps in a non-destructive manner. As a result, it’s important and determinant to recognize the parameters which affect the electrical conductivity and these parameters’ value of effect; and also to set some regulations out, to offer a mix design with a specified electrical conductivity. The existing models about describe the description of concrete behavior while passing electrical current, and the electrical conductivity prediction, are categorized in two divisions. The first category, is about to describe the behavior with respect of, each components (cement paste and aggregate) electrical conductivity and mix ratios. But in the second category, only the properties of conductive phase (cement paste or evaporable pore water of cement paste) are considered. In this research, studying the both categories, it is attempted to offer some models (using artificial neural network and regression correlations methods) to predict concrete electrical conductivity, based on the properties before mixture, and different parameters effects on concrete electrical conductivity are investigated through this. Considering this fact concretes have produced, cured and tested in two categories. The first set is specified for cement paste and aggregate investigations, including 80 mix designs with parameters of volume, type, sieve analysis and maximum size of aggregates, and their corresponding cement paste. The ages for testing samples electrical conductivity are 3, 7, 28 and 150 days. In addition to, the electrical resistivity of aggregates of these concretes are measured using an innovative method. The second set is specified for the evaporable water of cement paste and cement chemistry, includes 8 concrete mix designs with parameters of aggregate volume, cement type and w/c ratio, and their corresponding cement pastes are tested in 3 and 28 days. The second category tests include testing cement paste and concrete conductivity and measuring the volume of water-permeable pores and evaporable water extraction, analysis and measuring the electrical conductivity. Except this, a device is designed and manufactured to extract the evaporable water of cement paste. The results demonstrate that the concrete behavior description in passing electrical current and prediction of electrical conductivity using each components electrical conductivity had done precisely. In addition to this, the artificial neural networks architecture had carefully offered considering maximum size of aggregates, and electrical conductivity of aggregate and cement paste and their bulk ratios to predict the concrete electrical conductivity, and a regression correlation had offered to predict the concrete electrical conductivity using electrical conductivity of cement paste, aggregate and their bulk ratios. Except this, the investigations prove that fineness and decrease of maximum size of aggregates result in electrical conductivity increase, and the increasing the bulk ratio and electrical resistivity of aggregate would decrease the electrical conductivity of concrete. Key Words: Concrete electrical conductivity, Modeling, Artificial neural network, Regression, Cement paste, Aggregate, Evaporable cement paste pore water.