PurposeFailure of the materials occurs once the stress intensity factor (SIF) overtakes the material fracture toughness. At this level, the crack will grow rapidly resulting in unstable crack growth until a complete fracture happens. The SIF calculation of the materials can be conducted by experimental, theoretical and numerical techniques. Prediction of SIF is crucial to ensure safety life from the material failure. The aim of the simulation study is to evaluate the accuracy of SIF prediction using finite element analysis.Design/methodology/approachThe bootstrap resampling method is employed in S-version finite element model (S-FEM) to generate the random variables in this simulation analysis. The SIF analysis studies are promoted by bootstrap S-version Finite Element Model (BootstrapS-FEM). Virtual crack closure-integral method (VCCM) is an important concept to compute the energy release rate and SIF. The semielliptical crack shape is applied with different crack shape aspect ratio in this simulation analysis. The BootstrapS-FEM produces the prediction of SIFs for tension model.FindingsThe mean of BootstrapS-FEM is calculated from 100 samples by the resampling method. The bounds are computed based on the lower and upper bounds of the hundred samples of BootstrapS-FEM. The prediction of SIFs is validated with Newman–Raju solution and deterministic S-FEM within 95 percent confidence bounds. All possible values of SIF estimation by BootstrapS-FEM are plotted in a graph. The mean of the BootstrapS-FEM is referred to as point estimation. The Newman–Raju solution and deterministic S-FEM values are within the 95 percent confidence bounds. Thus, the BootstrapS-FEM is considered valid for the prediction with less than 6 percent of percentage error.Originality/valueThe bootstrap resampling method is employed in S-FEM to generate the random variables in this simulation analysis.

Ball screw is an essential mechanical component of computer numerical controlled (CNC) milling machine, which the positioning accuracy of ball screw is highly associated with lead angle accuracy and axial clearance. In particular, the inaccuracy of lead angle and axial clearance of ball screw can be massively affected the inaccuracy of positioning, leading to the degraded quality of manufactured products. Therefore, a reliable and productive measurement system analysis is indeed in monitoring the accuracy parameters of the ball screw. The main objective of this study is to propose using the Bootstrap simulator in monitoring the accuracy parameters of measurement system for ball screw, with the abstraction of cost and time. The accuracy parameters of the measurement system are including stability, bias and linearity. Based on the simulation results, it can be concluded that the Bootstrap simulator is more productive in monitoring the accuracy parameters of measurement system for ball screw compared to the Monte Carlo simulator. This is due to the Bootstrap simulator can be yielded a lower uncertainty of simulation compared to the Monte Carlo simulator. Furthermore, the Bootstrap simulator is also more advantages compared to the Monte Carlo simulator as this simulator can be carried out with small sample size of measurement data.