Probabilistic Buckling Analysis on Light Steel Compression Members

Cold-formed steel sections are thin-walled structural members with high width-to-thickness ratios, making them highly susceptible to initial geometric imperfections that may reduce their buckling capacity. Although SNI 8399:2017 specifies allowable limits for imperfections such as bow, twist, and camber, it does not provide a quantitative assessment of their influence on column strength reduction. This study aims to evaluate the effect of geometric imperfections on the compressive capacity of cold-formed steel columns using a probabilistic approach. The research adopts the geometric imperfection measurement technique used by Pariatmono (1994), in which circumferential data are transformed into linear coordinates. Imperfection data were collected at 6° intervals, yielding 60 measurement points for each specimen. Each imperfection profile was reconstructed through Fourier analysis, producing Fourier coefficients for 60 columns. These coefficients were then processed to obtain mean values and variations at ±10%, ±20%, ±30%, ±40%, and ±50% of the standard deviation, resulting in 11 imperfection models. All models were subsequently analyzed using nonlinear buckling analysis in ANSYS Workbench 2022 R1 to determine their influence on compressive behavior. The results indicate that geometric imperfections significantly affect the buckling capacity of cold-formed steel columns. However, the available probabilistic data are insufficient to establish generalized imperfection limits for design purposes. Further studies with larger datasets are required to develop more representative tolerances.