Statistical analysis of tensile strength and flexural strength data from universal testing machine

• Main Authors: Shamsuri, Ahmad Adlie, Mat Darus, Siti Atiqa Al Zahra
• Format: Article
• Published: Sciencedomain International 2020
• Online Access: http://psasir.upm.edu.my/id/eprint/85903/; https://journalajpas.com/index.php/AJPAS/article/view/172
• Institution: Universiti Putra Malaysia

The tensile strength and flexural strength are the most important mechanical properties as they provide the value of maximum tensile stress and flexural stress. The objective of this study is to analyze statistically the tensile strength and flexural strength data obtained from a universal testing machine. The tests were conducted upon a thermoplastic, specifically high-density polyethylene (HDPE), which was in-house molded by using an injection-molding machine. Three different persons have performed the tensile and flexural tests. Three other laboratories have also been involved in these tests. The relative standard deviation (RSD) values were calculated to express the precision and repeatability of the tests. Later, the standard score (z-score) values were also calculated to aid the comparison of the data. Finally, the single-factor analysis of variance (ANOVA) was employed to investigate statistically significant differences between the means of the tensile strength and flexural strength data of each person and laboratory. From the calculation, the RSD values of all three persons and laboratories were lower than 5%, indicating that the data were consistent. The z-score values of all three persons were within the range from -2 to 2, suggesting that the data were close to average. However, the z-score value for one of three laboratories was not within the range, demonstrating that the data was unusual. The P-values of all three persons were higher than 0.05 (except for flexural strength), implying that the difference between the means of the data was not statistically significant. Nevertheless, the P-values of all three laboratories were lower than 0.05, indicating that the difference between the means of the data was statistically significant.