Purpose Keeping in view the diabetes status that has affected about 415 million people globally and is the leading cause of death in many countries along with therising demand for low Glycemic Index (GI) foods, the purpose of this paper is to optimize the extrusion process for the development of low GI snacks from underutilized crops like water chestnut and barley. Design/methodology/approach The extrusion parameters (screw speed and barrel temperature), feed moisture and water chestnut flour, barley flour proportion, were varied and their effects on system and product responses (specific mechanical energy, water absorption index, water solubility index, bulk density, expansion ratio and breaking strength) were studied. Findings All the system and product responses were significantly affected by independent variables. Response surface and regression models were established to determine the responses as function of process variables. Models obtained were highly significant with high coefficient of determination (R2=0.88). The optimum processing conditions obtained by numerical optimization for the development of snacks were 90°C barrel temperature, 300 rpm screw speed, 14 per cent feed moisture and WCF-to-BF ratio as 90:10. Shelf life studies confirmed that the developed snacks can be safely stored in HDPE bags for a period of six months under ambient conditions. Originality/value Water chestnut and barley flour did not blend till date for extrusion cooking. Such snacks shall be a viable food option for diabetic people and can act as laxative due to high fibre and β-glucan content from barley.

No abstract

No abstract

In the present study we have proposed an improved family of estimators for estimation of population mean using the auxiliary information of median, quartile deviation, Gini's mean difference, Downton's Method, Probability Weighted Moments and their linear combinations with correlation coefficient and coefficient of variation. The performance of the proposed family of estimators is analysed by mean square error and bias and compared with the existing estimators in the literature. By this comparison we conclude that our proposed family of estimators is more efficient than the existing estimators. To support the theoretical results, we also provide the empirical study.

No abstract