It is necessary to look at wheat productivity in the light of global food security, optimal utilization of inputs and formulating policies in view of the prevailing conditions. The present paper deals with advanced understanding of wheat yield forecasting through the use of process based crop models, machine algorithms and their hybrid approaches. This study considers the benefits and limitations of these forecasting methods in terms of climate change, soil characteristics or even self-forecasting. The review aims at enhancing effective yield prediction and in the process pinpoint areas which have not been covered in the existing literature for further investigation.