Comparative Analysis of ANN-MLP, ANFIS-ACOR and MLR Modeling Approaches for Estimation of Bending Strength of Glulam

Multiple linear regression (MLR), adaptive network-based fuzzy inference system–ant colony optimization algorithm hybrid (ANFIS-ACOR) and artificial neural network–multilayer perceptron (ANN-MLP) were tested to model the bending strength of Glulam (glue-laminated timber) manufactured with a plane tree (Platanus orientalis L.) wood layer adhered different weight ratios (WR) modified starch/urea formaldehyde (UF) adhesive containing levels nano-ZnO (NC) used at press temperature (Tem) time (Tim). According X-ray diffraction (XRD) stress–strain curves, some changes in behavior product seen. After selecting best through determining statistics such as determination coefficient (R2) root mean square error (RMSE), absolute (MAE) sum squares (SSE), production process was optimized obtain highest modulus rupture (MOR) using Genetic Algorithm (GA) combined MLP. It determined that MLP had accuracy estimating response. MLP-GA hybrid, optimum input values for obtaining response include: WR—49.1%, NC—3.385%, Tem—199.4 °C Tim—19.974 min.