A decision-support system that allocates concentrated feed to cows through
individual feeders according to their performance has been developed using
fuzzy logic. Fuzzy sets were constructed by analysing experimental data bas
ed on which ranges of values for the fuzzy sets members were determined by
an expert. The model was developed on a group of 15 cows in a dairy of 50 m
ilking cows (two milkings a day) under a controlled dietary regime, assumin
g characteristic cow behaviour concerning body weight, milk yield, lactatio
n period, and concentrate feed consumption. The data were analysed by the d
ecision-support system and compared to weekly decisions of the expert. Resu
lts indicated relationship between historical cow performances and concentr
ated food intake. The advantage of using fuzzy logic was that it enabled im
proved concentrate feeding ration according to performance. The system enab
led to automate decision making, thus providing the farmer with a valuable
tool. However, from an economic point of view no significant statistical im
provement was achieved by the fuzzy logic system. (C) 2001 Silsoe Research
Institute.