Optimization of mining projects is often aimed at maximizing the net present value (NPV). Cut-off grade along with production rate determines the quantity and destination of material that is mined and processed. Thus, the cash flows and the NPV of a mining project are directly affected by the cut-off grade, the mineable reserve and the production rate. In order to achieve the maximum NPV, these factors must be evaluated. Block caving is a non-selective mass mining method. In block caving method, as the cut-off grade changes, the amount of mineable reserve, and the correlated mining envelope changes consequently. Determining the optimum cut-off grade and production rate for block cave mining is a complex task, therefore, artificial neural network (ANN) and response surface method (RSM) approaches are utilized in this paper. According to the results, a combination of RSM and ANN models is able to determine the best configuration of cut-off grade and production rate that leads to the maximum NPV.