| Sumario: | Zhero is an indigenous popular snack product of Bhutia and Tamang tribe of Sikkim, India. The present study focusses on standardizing its process operations such as soaking and grinding for production of uniform quality rice flour to be used as raw material in Zhero preparation for commercial processing. The effects of soaking time, St (4 – 8 h) and temperature, ST (15-25 °C) and grinding time, GT (1 – 2 minutes) on the characteristics of rice flour have been studied. The process parameters were optimized using a multi-objective genetic algorithm (GA) along with a three-layer feed forward artificial neural network (ANN). The ANN model developed could satisfactorily predict all responses with R value of 0.999 for all training, testing, validation and global sets of data and MSE value 0.01883 for validated dataset. The final optimum conditions (St: ST: GT) were 4.2 h, 18 ℃, and 1 min respectively. The predicted rice flour quality at optimized process conditions were: 23.85 % w b moisture content, 0.543 g/cm³ bulk density, 0.57 g/cm³ tapped density, 28.09° angle of repose, 0.31 mm particle size, Carr’s index 5.055 and HR ratio 1.05. The experimental and ANN-GA model predicted responses had a relative percent error of < 10%, suggesting suitability of the developed model. This study represents a new attempt to conduct appropriate scientific research to determine the consistency of the production process and the quality of the final product, not only to confirm its origin and maintain its culture, but also to improve and standardize its technology for future commercial success.
|
|---|