Vol. 10, Issue 6, Part D (2024)

Abstract
The objective of the present study was to apply an artificial neural network (ANN) to investigate the relationship between energy inputs and mustard production in the South Kashmir region of Western Himalayas, India. The energy consumption pattern was determined for this purpose by collecting data from 132 farmers using face-to-face questionnaires. Human work, chemical fertilizer, farmyard manure (FYM), seeds, machinery, and diesel fuel all contribute to the total energy input. The total input and production energy for mustard output were found to be 3101.04 and 8552.49 MJ acre^-1, respectively. The energy indices for energy use efficiency, energy productivity, specific energy, and net energy were estimated to be 2.7, 0.11 kg MJ^-1, 9.06 MJ kg^-1, and 5451.44 MJ acre^-1, respectively. The findings found that the total consumed energy input may be classified as direct energy (29%) and indirect energy (71%) or renewable energy (33%) and non-renewable energy (71%). The lowest RMSE and MAPE of 0.0704 and 0.57 were obtained at 10 neurons, according to the modeling implementations. The best forecasting model was found at the equal amount of neurons. For mustard yield, the best topology's coefficients of determination (R²) were 0.90. The generated ANN's promising potential reveals that it is a powerful and stout instrument that can be used an efficient and dynamic field of research in the arena of energy usage prediction.