Abstract

Leather’s handle and character has been evaluated and classified using an artificial neural network (ANN). Experimental mechanical parameters correlated to elasticity, softness, fullness and extensibility were obtained. The feed-forward neural network used provided accurate modelling between the input parameters, bend constant (Kb); compression index (Kc); lastometer coefficient (Kl); attenuation constant (αb); attenuation speed constant (K´) and the output parameter (leather’s handle and graded character). Although only a small number of experimental data were used in training the neural network, the results obtained are comparable to the results of evaluations done by the traditional method.