Artificial neural network evaluation of cement-bonded particle board produced from red iron wood (Lophira alata) sawdust and palm kernel shell residues

As a way of promoting environmental sustainability, it becomes paramount to salvage the quantity of agricultural wastes being destroyed or disposed into the environment. A novel strategy to reduce these wastes is by reusing them. In the present study, the physical and mechanical properties of particleboards produced from red iron wood (Lophira alata) sawdust and palm kernel shell (PKS) was evaluated by artificial neural network (ANN). The production of this particle boards involved the synergistic combination of effective parameters such as percentage composition of cement, sawdust and palm kernel shell varied between 25-40, 20-50 and 20-50 respectively. The boards were tested for physical properties such as water absorption (WA), thickness swelling (TS), density and mechanical properties such as modulus of rupture (MOR) and modulus of elasticity (MOE). The networks was trained and tested by Multilayer Normal Feed Forward Perceptron (MNFFP), with a quick propagation learning algorithm. The performance of the ANN network shows it has a high potential for predicting the properties of cement bonded particle board.