PREVENTIVE DIAGNOSIS OF LAMENESS IN DAIRY COWS

This research aimed to develop a Fuzzy inference based expert system to help preventing lameness in dairy cows. Hoof length, nutrition parameters and floor material properties (roughness) were used to build the Fuzzy inference system. The expert system architecture was defined using Unified Modelling Language (UML). Data were collected in a commercial dairy herd using two different subgroups (H1 and H2), in order to validate the Fuzzy inference functions. The number of responses True Positive (TP), False Positive (FP), True Negative (TN) and False Negative (FN), were used to build up the systems classifier, after comparison with an established gold standard. A Lesion Incidence Possibility (LIP) developed function indicates the chance the cow to become lame. The actual hoof lameness percentage in the H1 was 8.40%, and in the H2 was 1.77 %. The results reached, estimated a Lesion Incidence Possibility (LIP) of 5.00% in H1, and a LIP of 2.00 % in H2. The simulation using the system in H1 departed 3.40% from the actual lameness data, while in H2 the difference between simulation and actual data was 0.23% indicating the efficiency of the decision making system.