As a new solid-state joining process, friction stir welding (FSW) has been a prominent technique in welding aluminum alloys. Its applications are soaring in the automotive, shipbuilding, railroad, and aerospace industries. Similar to any added-value and novel manufacturing process, the FSW process has undergone extensive research and developments in diverse inter-disciplinary areas. The major research and development issues can be categorized as: (1) product enhancement and (2) process development. Researchers attempt to modify and promote the FSW product properties through microstructural studies, mechanical properties, material flow, thermal, structural, and residual stress analysis. The process developments, conversely, encompass the subjects related to the FSW tool design, process setup, machine design, and eventually the process control and automation. In order to address the significant demands of the industry, in this dissertation, a multi-disciplinary study of the FSW process is covered to describe: (i) numerical modeling of the material flow and heat transfer, (ii) process monitoring, and (iii) process parameter optimization. In the numerical modeling of the process the attempts are placed as a multi-physics problem on the concise three-dimensional (3-D) computational fluid dynamics (CFD) modeling of the FSW. The preparation for the physical model, problem assumptions, governing equations, and the CAD model preparation and meshing are discussed. In the post-processing, the goal is to evaluate the loads carried by the tool and their variation by the welding parameters. The influence of the rake angle on the variation of the loads carried by the FSW tool and on the generation of defects is studied. Furthermore, the effect of the process parameters on the initiation of the channel-like defects in the weld is discussed using the numerical modeling and the microstructural studies methods. In the process monitoring, the acoustic emission (AE) monitoring technique is employed to acquire the distinct valuable events arising in the FSW process. The study is composed of (1) monitoring the friction stir (FS) butt welding of aluminum alloy (Al6061-T6), (2) monitoring the FS butt welding of dissimilar materials (AISI 1018 -- Al6061-T6), and (3) monitoring the initiation of the channel-like defects by studying the axial force and AE signal. Optimizing the friction stir spot welding (FSSW) process---a derivative of the FSW process---is the subject of our final study in this dissertation. In this study, a genetically-optimized neural network system (GONNS) is introduced to simulate the FSSW process and optimize the process parameters based on the objectives of the user.