Wood Polymer Composites Optimizer with Genetic Algorithm

Wood polymer composites, or natural fiber composites are basically made up from a combination of plastic and wood flour. Usually, its quality is measured and depended on the mechanical properties that were used as composite materials for product development. Hence, a lot of laboratory experiments were conducted in factories for testing the durability of this product. However, there is still a lack of the solution for an automated system to formulate the wood polymer composite composition efficiently. This paper addresses the use of a genetic algorithm in finding an optimal composition for each of the properties used in producing a better-quality product. A new solution representation to map the composite composition is pro-posed. Different parameters were employed in the computational experiments with the aim to find the best fitness value for the composition. The stochastic behavior that was embedded in genetic algorithm has demonstrated diverse solutions with regards to the search exploration and exploitation capability towards the feasible solutions and convergence. The findings demonstrate a great potential ability of genetic algorithm in finding good accuracy for other wood polymer composites such as tropical timber as wood polymer materials.  It is expected to reduce cost and time as compared to traditional laboratory way.