Abstract Artificial intelligence is playing a vital role in oil and gas industry. In this paper a new procedure and a unique statistical model has been developed to predict critical oil rate at any hydraulic fracture perimeter and reservoir property for Tight Oil Reservoirs (TOR) containing hydraulic fracture under water coning issue. The new approach uses numerical simulation and Artificial Neural Networks (ANN). Simulating the coning behavior is an expensive and time consuming technique. Therefore, there is a need for a readymade correlation which can help as a good quick estimate. This research work exhibits a 3D simulation model, which has run for different ranges of fracture conductivity, kf*w (md-ft), reservoir permeability, k (mD), anisotropy ratio, kv/k, density difference, ?w-?o (lb/ft3), water oil contact, WOC (ft), fracture length, Lf (ft), fracture height, hf (ft) and oil viscosity, µo (cp) to find critical oil rate. Total number of 20,000 data points are obtained by the 20,000 number of simulation runs by coupling the commercial reservoir simulator with MATLAB. ANN is employed to obtain generalized correlation for critical oil rate in hydraulically fractured tight oil reservoirs. Seventy percent of the simulated data points are used to develop the correlation and rest of them are used to validate and test the proposed correlation. Results show the good agreement with the unseen data. Correlation parameters are transformed to dimensionless form in order to get generalized correlation. Nonlinear regression technique was also applied to find the correlation, but ANN gives good results with lesser error as compared to non-linear regression. It is observed that critical oil rate highly depends on fracture half length, fracture height and fracture conductivity in the hydraulic fractured vertical oil well.