The use of welding processes, especially for joining of aerospace alloys has gained a significant importance in the recent years. This is owing to the enhanced joint efficiency, increased sealing ability and reduced weight of the welded structures as compared to riveted structures. Moreover, the modern trend in aerospace industry has shifted towards the use of titanium alloys, due to their high strength to weight ratio and good corrosion resistance. This work is focused on the welding of the well-known α titanium alloy Ti-5Al-2.5Sn, which haslow cost alloying elements as compared to the mostly widely used Ti-6Al-4V alloy, has a good weldability and is also more suitable for high temperature aerospace applications. Tungsten inert gas (TIG), laser beam welding (LBW) and electron beam welding (EBW) are the mostly used welding methods for titanium alloys. As compared to TIG welding, LBW and EBW are always the preferred welding methods due to low heat input and deep penetration characteristics. However, TIG welding is mostly employed industrially due to significantly less capital cost and ease of automation due to reduced equipment size. A number of gaps were identified in the open literature related to the welding of Ti-5Al-2.5Sn alloy. Firstly, few studies are available in the public domain related specifically to the welding of Ti-5Al-2.5Sn alloy using TIG, LBW and EBW. Moreover, the reported work related to comparison of TIG, LBW and EBW of other titanium alloys is limited and there is a need of in-depth, comprehensive comparison of these welding processes in terms of microstructure, mechanical properties and residual stresses in the welded structures. The opportunities available for parametric analysis of LBW process in titanium alloys and optimization of the pulsed TIG welding process for titanium alloys especially Ti-5Al-2.5Sn alloy have not been explored to full potential. The present work aims mainly at improving the pulsed TIG (P-TIG) welding process for 1.6 mm thick Ti-5Al-2.5Sn alloy sheet so that resultant microstructure, mechanical properties and residual stresses are comparable to that of pulsed LBW (P-LBW) and EBW weldments. Microstructure, oxide contents and microhardness of fusion zone, HAZ width, weld zone strength, tensile residual stresses and plate deformations were measured to compare the performance of the weldments. P-LBW was found to be most suitable in terms of these performance attributes of TI-5Al-2.5Sn welds due to low heat input which led to a complete martensitic transformation in the FZ. The absence of shielding gas due to vacuum environment in EBW was beneficial in terms of increasing the joint quality (low oxide contents). However, an increased width of heat affected zone (HAZ) and partial α’ martensitic transformation in FZ of EBW was observed as compared to P-LBW. High heat input and much wide heat source in P-TIG led to coarse microstructure and partial martensitic transformation in FZ resulting in increase of FZ and HAZ width, plate deformations and tensile residual stresses and a reduction in FZ microhardness and weld zone strength. The optimization of P-TIG welding was performed using Box-Behnken design of experiments in which a mathematical was developed to establish the relation between the welding input factors (peak current, background current and welding speed) and output responses (FZ width, HAZ width, FZ grain size, ultimate tensile strength, notch tensile and impact strength, and elongation, longitudinal and transverse residual stresses). The dependence of output responses on the inputs of P-TIG welding and its physical significance in the context of microstructure was discussed in detail. Optimization was performed through different criteria and a multi-response optimization was suggested to maximize the joint strength, impact properties and minimize the residual stresses. Results were experimentally validated and the range of welding input parameters were recommended through overlay plots for industrial application.