The primary objective of the present work is to introduce a new method, i.e., Theory of Experimentation for prediction of monthly average global solar radiation. Meteorological data for 15 years is accessed considering six input predictors (i.e., latitude, longitude, altitude, relative humidity, temperature, and sunshine hours). Global solar radiation model is developed using various input parameters, and the accuracy of the developed models is assessed using statistical errors. The established model forms are also compared with the models available in the literature. Also, Global Performance Indicator is employed to sort the models for the development of the ranking system. A five-variable global solar radiation model (M-06) is found the best amongst all the proposed models (on training dataset) where the determination coefficient is 0.9424, and the mean percentage error is -0.1524%; whereas, for validation dataset, a two-variable regression model was seen to be the best. The study reveals that the effectiveness of the developed Global solar radiation model does not increase with an increase in the input variables; however, altitude, relative humidity, and sunshine hours are the dominating parameter. The proposed method exhibits a high potential of use in the prediction of monthly average global solar radiation. (C) 2019 Elsevier Ltd. All rights reserved.