Laser powder bed fusion additive manufacturing (PBF-LB AM) has matured from rapid prototyping into an industrial manufacturing technology used to produce critical end-use parts for demanding applications in the aerospace, nuclear, and other regulated industries. However, the high variability in the mechanical properties of parts produced using PBF-LB AM is a major hindrance to further advancement and wider adoption of this potentially revolutionary manufacturing technology.

      The purpose of this dissertation is to identify the root causes of the observed variability in properties of PBF-LB manufactured parts and quantify their impact. This is explored in the context of machine architecture-defined process parameters, which have rarely been adequately considered in studies related to PBF-LB material properties. Specifically, the objective is to study the effects of shielding gas flow speed, re-coater blade type, and laser beam focal point position on the quality of PBF-LB manufactured parts. Practical guidance is provided on how these parameters should be considered in additive manufacturing procedure specifications (AMPS) or similar process control approaches aimed at assuring repeatable material properties. Furthermore, the variability in mechanical properties of 316L stainless steel produced using different machines and powders was studied, including exploration of the potential of using standardized post-process heat treatments to reduce variability. Additionally, this thesis presents strategies for in-situ process monitoring for quality assurance purposes.

      The results highlight the significant impact of the previously understudied machine architecture-defined process parameters on the PBF-LB process. These parameters should be considered essential to the process. Post-processing with hot isostatic pressing reduced variability in most of the studied properties, but at the cost of an associated significant reduction in absolute properties. Co-axial photodiode-based melt pool monitoring and contact-image sensor-based powder bed imaging are effective means of direct and continuous monitoring of the state of the actual fundamental unit processes in the PBF-LB process: the spreading of the powder layer and the selective laser melting of it.