Teke, Ibrahim T.Ertas, Ahmet H.2026-02-082026-02-0820252228-61872364-1835https://doi.org/10.1007/s40997-025-00859-1https://hdl.handle.net/20.500.12885/5595This study investigates the fatigue behavior of spot-welded tensile shear (TS) joints, emphasizing the influence of geometric parameters on fatigue performance under variable cyclic loading. Experimental fatigue testing of TS specimens reveals distinct failure mechanisms dependent on applied load levels. At high loads, failure predominantly occurs through shear at the weld nugget, whereas lower loads lead to crack propagation into the surrounding base material. To quantitatively assess the role of geometric parameters-including plate width, length, thickness, and overlap distance-a regression-based predictive model was developed, utilizing Delta F-N curves. The results indicate that increased plate width and thickness enhance fatigue resistance up to critical stabilization thresholds, while variations in plate length exhibit a non-monotonic influence, initially reducing fatigue strength before reaching an equilibrium. Overlap length, however, demonstrates minimal impact on fatigue life. Furthermore, the regression model, originally formulated for single-lap joints, was successfully applied to TS specimens, demonstrating its robustness and predictive accuracy across different joint configurations. These findings underscore the critical role of geometric optimization in enhancing the fatigue resistance of spot-welded structures and provide a robust framework for predictive modeling across various fatigue regimes, ensuring reliability under diverse cyclic loading conditions.eninfo:eu-repo/semantics/closedAccessFatigue behaviorSpot-weld jointsGeometric parametersVariable load conditionsRegression modelingSubmodelingArticle10.1007/s40997-025-00859-149416191647WOS:0015102954000012-s2.0-105008324496Q3Q2