The increased use of integral structures is critical for developing improved commercial transport structures and unmanned and manned space transport. They reduce the use of rivets and the need to create lap joints, the primary source location of multi-site fatigue damage, especially in airframe structures. One method being examined is to produce and join large modular units by welding. Welding is essential in the construction, aircraft, aerospace, and automotive industries. However, welded joints are sensitive to fatigue failure due to cyclic loading and fatigue crack propagation influenced by welding residual stresses (RS) distribution. Welding residual stresses are caused by differential thermal expansion and contraction of the weld metal and parent material. In and near the weld, residual stress levels can be very high, up to material yield strength magnitude in highly constrained situations, which is the case in most real structures. The residual stress in a welded joint can augment the externally applied load and cause structural failure. The fatigue life of the welding can be increased by machining the weld bead to reduce from higher to lower tensile residual stress, and compressive residual stresses in the weld bead are removed. This research analyzes the fatigue strength developed by a robotic welding process and subsequent machining of the weld bead through a control numerical process in duplex stainless steel 2205 (DSS). Two main variables will be studied: i) heat input as a welding variable and ii) chip cross-sectional area as a machining variable. Results showed that X-ray diffraction measurement along the center thickness of the machined welded plates reveals compressive residual stress near the adjacent weld region, and the fatigue life of the machined weldment are enhanced by 80% at a stress level of 550 MPa and 92% at 350 MPa compared to the DSS welds with the weld bead.