The crashworthy ability of glass/epoxy (GFRP) square tubes with induced holes was studied under quasi‐static axial crashing. Four design parameters were chosen, each at three levels, to calculate crashworthiness indicators. The design variables are hole diameter (d), hole position/specimen length (P/L), number of holes/distribution (n), and speed (V). To get the optimum crashworthiness parameters, the design of experiments approach using Taguchi technique was adapted. Optimal conditions with minimum initial peak force (normalFip), maximum absorbed energy (U) and maximum specific energy absorption (SEA) were determined. Main effect, signal/noise ratio (S/N), and analysis of variance (ANOVA) were determined using MINITAB 18. Experiments based on L9 orthogonal array were performed. Results indicated that the dominant influencing parameter on normalFip is “n” with a contribution of 93.97%. “V” followed by “d” are the highest influencing parameters on “U” with a contribution of 49.07 and 43.02%, respectively. “d” is the highest influencing parameter on SEA, followed by “V” with a contribution of 50.71% and 46.12%, respectively. Confirmation tests were carried out to validate the estimated model with respect to experimental results. normalFip, U, and SEA of intact tube were 45.05 kN, 830.32 × 10−3 kJ, and 8.23 × 10−3 kJ/g, respectively. The optimum normalFipof GFRP tubes with cutouts was found to be 38.59% smaller than that of the intact specimen. The optimum U and SEA of are, respectively, 4.73% and 25.76% greater than those of intact specimen.