Effect of fibre-quality parameters on pulp properties by using multiple linear regression and artificial neural network.

This article investigates the relationship between changes in pulp fibre-quality parameters caused by refining and the resulting paper-making properties, using multiple linear regression (MLR) and artificial neural network (ANN) analyses. The study analyzes fibre characteristics—including fine content, fibre length and width, degree of external fibrillation, curl index, kink index, and coarseness—in pulps derived from rice straw, hardwood (Trema orientalis and Eucalyptus camaldulensis), and softwood, correlating these with tensile, tear, and burst indices of paper sheets. Results show that MLR models effectively predict these paper strength properties (adjusted R² values of 0.85–0.97), while ANN models provide even higher predictive accuracy (R² > 0.95) for most cases, particularly for rice straw and softwood pulps. The study highlights the kink index as the most influential fibre parameter affecting paper strength and demonstrates the potential of ANN as a robust tool for optimizing pulp refining processes based on fibre morphology.