Investigation on the usability of some mathematical models in in vitro gas production techniques

Abstract

The aim of this study was to investigate the usage of some mathematical models in in vitro gas production techniques. With this aim, the Logistic, Orskov, Verhulst, Janoscheck, Weibull, Bridges, Mitscherling, Monomolecular and Von Bertalanffy models were used, respectively. The goodness of fit of these models to in vitro gas production data was examined by using various criteria, such as Mean Square Error (MSE), Adjusted coefficient of determination (¯R2 ), Accuracy factor (AF) and Bayesian Information Criterion (BIC). However, autocorrelation and the distribution of residuals were examined with the Durbin Watson and Shapiro Wilks tests, respectively. Although the Verhulst and Logistic models showed a lower goodness of fit according to the other models, these models were found to be suitable for in vitro gas production studies in the result of all the criteria and tests. As a result, it was determined to be suitable for the in vitro gas production technique of other models except for the Orskov model.