Comparison of Genetic Algorithm and Taguchi Optimization Techniques for Surface Roughness of Natural Fiber-Reinforced Polymer Composites

Climate change has necessitated the development of “green” alternatives to replace existing materials. This focus has resulted in the push toward fabricating natural fiber-reinforced polymer composites. This research work looks at the surface roughness (SR) of natural fibers like rice husk ash (RHA) and groundnut shell ash (GSA) reinforced in nine different concentrations into an epoxy matrix to form composites. Composite samples are fabricated using various concentrations of natural fibers and measures and optimizes for the SR through the implementation of genetic algorithms (GA). It was found that a minimum SR of 1.422 μm can be obtained for an epoxy/hardener ratio of 3:1 and without the addition of any reinforcements. This optimization was achieved within 102 generations. In addition to GA optimization, another optimization implementation was done through the Taguchi method. An array referred to as the L9 orthogonal array was constructed with epoxy/hardener ratio, as well as the weight percentage of RHA and GSA taken as the input parameters. The mean effects plot for means and the main effects plot for signal-to-noise (S/N) ratio were plotted according to the smaller-the-better criterion, since the output variable was SR. It was found that the results obtained from the Taguchi method closely agreed with that of GA, that is, minimum SR was found for an epoxy/hardener ratio of 3:1 and 0% addition of reinforcements. Analysis of variance (ANOVA) helps us figure out that the epoxy/hardener ratio was the most important factor affecting the response variable accounting for about 36.35% of the total effect.