Seaweed drying characterization via serial statistical criteria analysis

Abstract

Abstract In drying seaweeds, it is important to describe its drying behavior using semi-theoretical models. Predicting the best drying model for the seaweed was commonly done by testing the goodness of the curve for seaweed drying models by calculating the coefficient of determination R2, the reduced chi square X 2, root means square error (RMSE) and mean bias error (MBE). However, it is essential to rank the drying models before selection by assessing its performance index Φ and mean relative deviation (MRD). It is also necessary to test the normality by evaluating the skewness and kurtosis statistics using the D’ Agostino Pearson test which is the best test for goodness of fit. The simple sample run test statistics and confidence interval must also be satisfied These are the conditions that should be met under the serial statistical criteria analysis. This study utilized it to choose the best drying model for seaweeds. Employing this statistical analysis for testing commonly used drying models, Modified Page model emerged as the best model that depicts the drying behavior of the seaweeds with accuracy of 99.98% and highest performance index of 452.1967.