Milk Production Forecasting by a Neuro-Fuzzy Model
Pp. 3-11 (9)
Atsalakis S. George, Parasyri G. Maria and Zopounidis D. Constantinos
Many fields are increasingly applying Neuro-fuzzy techniques such as in model
identification and forecasting of linear and non-linear systems. This chapter presents a neuro-fuzzy
model for forecasting milk production of two producers. The model utilizes a time series of daily data.
The milk forecasting model is based on Adaptive Neural Fuzzy Inference System (ANFIS). ANFIS uses
a hybrid learning technique that combines the least-squares method and the back propagation gradient
descent method to estimate the optimal milk forecast parameters. The results indicate the superiority of
ANFIS model when compared with two conventional models: an Autoregressive (AR) and an
Autoregressive Moving Average model (ARMA).
Milk forecasting, neuro-fuzzy, ANFIS, AR, ARMA, forecasting, milk production.
Department of Production Engineering and Management, Technical University of Crete, 73100 Chania, Crete, Greece