Statistical inferences in material selection of a polymer matrix for natural fiber composites

Abstract

In this paper, statistical inferences in material selection of polymer matrix for natural fiber composite are presented. Hypothesis testing and confidence interval were used to evaluate the suitability of the sample for use as a matrix in natural fiber reinforced composites. The screening process for material selection was carried out using a stepwise regression method. Then, the ranking process in material selection was conducted using an estimation of performance score (PS) for mechanical properties such as impact strength (IS), elongation at break (E) and tensile strength (TS). Ten types of polymer were involved in the study. The final selection revealed that polyamide (PA6), polyurethanes (PUR) and polypropylene (PP) are the potential candidates to manufacture handbrake levers according to IS, E and TS, respectively. Here, it was found that the score for Tp (thermoplastic) is better than Ts (thermoset) in terms of IS. In contrast, the Ts offered a better score result than, Tp, with respect to E and TS. The results of statistical measurements using statistical modelling prove that the data analysis can be used as a part of the decision making in material selection.