Analysis and Prediction of Glove Quality Based on Manufacturing Factors

This paper considers the effects of manufacturing factors on glove quality. Four factors are considered, namely the curing temperature profile, temperature of latex, percentage of calcium nitrate and the oven temperature after coagulant dip. The glove quality is measured based on the weight of the gloves, tensile strength after ageing, number of gloves with pinholes and finger thickness. The analysis is carried out using Wilcoxon signed rank tests where the significance of each factor on the glove quality can be determined. The glove quality is subsequently predicted using neural networks. It is shown that the weight of the gloves can be predicted quite well using a simple neural network structure. However, the other quality measures cannot be predicted equally well. More complicated input-output relationships are likely to exist for these. The results from this work are important for aiding decisions on process changes in glove manufacturing.