Comparative assessments of multivariate nonlinear fuzzy regression techniques for egg production curve

The modelling process of egg production curves, where environmental and genetic factors are highly effective, is quite complex and difficult. In particular, the limitations of measurement and the errors encountered during the measurement process may cause uncertainty in the egg production process. In this study, multivariate nonlinear fuzzy regression analysis was used by configuring neural networks and least squares support vector machines in order to express the uncertainty in the system structure during the egg production process. This method was used to obtain the predicted values for egg production in the fuzzy frame. In the study, two different data sets were used which were measured for egg performance and egg weight variables in daily and weekly time periods. Multivariate nonlinear fuzzy regression analysis results were compared with both the observed values and the multivariate classical regression analysis results. Results of analysis show that multivariate nonlinear fuzzy regression analysis with neural networks is more successful than other methods and can be used as an alternative to classical methods in poultry farming.