Cost is one of the major factors in decision making at the early stages of a building construction process. The real challenge in cost estimation of building projects at the early stage is lack of information. So decision making is important in cost estimation for building design processes which needs an estimation tool for designers, estimators and project managers. Artificial neural networks (ANN) method is most effective and appropriate technique for initial stage cost estimation. This project is highlighting the study of Application of Artificial neural network (ANN) for pre design cost estimation of building projects to investigate and overcome problems caused in estimating project cost at pre-design stage of building projects. As well as to develop & test a Graphical User Interface (GUI) model of cost estimating for building projects in the early design phase using MATLAB software. Twelve actual real-life cases of building projects constructed in Pune District during the Three year period 2014-2017 were used as training materials. The architecture of Artificial neural network is presented for the estimation of the project cost at the initial stage.