The Double KT (DKT) multi-planar tubular joint is frequently found in the jacket offshore structures. An important aspect in designing the multi-planar tubular joint is accuracy in predicting the Stress Concentration Factor (SCF). Despite its practicability with considerable accuracy on stress analysis and estimating the SCF, the Finite Element (FE) method needs high computational time and effort. Therefore, this study will develop alternative SCF equations for DKT tubular joint with regression analysis as one of the machine learning techniques to increase equation's accuracy while reducing computational time. The variation of the DKT tubular joint was determined based on the validity range of the geometric parameters of the tubular joint (β, τ , and γ) and the combination of axial, in-plane bending (IPB), and outplane bending (OPB) moment loadings. Stress distribution and concentration factor of the DKT joints were analyzed based on FE model of the joint. Then, the SCF results from the FE analysis were used in the regression analysis to obtain new equations. Six SCF equations including for both brace-side and chord-side have been obtained while the reliability of the equation has been checked using Acceptance Criteria based on UK Department of Energy and showed good results.

It is important to consider geometric parameters on Stress Concentration Factor (SCF) at the tubular joint, especially multi-planar joint, when designing a jacket platform. Thus, a jacket platform capable to withstand the extreme environmental loads. Many researchers use the Finite Element (FE) method to analyze stress at joints and estimate the SCF with the variation of geometric parameters. That studies obtained the result that geometric parameters have significant effect on decreasing or increasing the SCF. Therefore, this study will delve the effect of geometric parameters (β, τ, and γ) on maximum SCF caused by axial loading, in-plane bending moment, and out-of-plane bending moment at DKT tubular joint. The increase of β results in the decrease in maximum SCF at brace-side and chord-side weld toe, respectively. The increase of τ and γ results in the increase in maximum SCF at brace-side and chord-side weld toe, respectively.