This paper employs fracture mechanics to perform damage tolerance analysis of Additively Manufacturing (AM) parts. The motivation of this study is that AM parts can be prone to fatigue and fracture failure due to porosities induced by the AM process. However, many parts are not being designed to be flaw tolerant. Implementation of damage tolerance requirements can be used to increase the robustness of designs and to reduce the likelihood of failure due to flaws, especially under cyclic loads. Applying damage tolerant design principles to AM components is an active research field. Challenges in applying fracture mechanics to AM parts include lack of established fracture properties of AM materials and porosities unique to the AM process. Depending on process parameters such as laser power and scanning speed combination, keyhole porosity or lack of fusion porosity can manifest in AM parts. Due to different forming mechanisms, keyhole and lack of fusion porosity can have significantly different shape and size that govern crack growth. A literature review of damage tolerance analysis of AM parts will be provided in the final paper. Several types of flaws unique to the AM process will be studied: surface flaws due to surface roughness inherent to the AM process; spherical volumetric flaws due to keyhole porosity or entrapped gas; irregular flaws due to lack of fusion porosity. Near surface volumetric flaws that can grow and transit into large surface flaws, which can cause accelerated crack growth and rapid failure; Large aspect ratio flaws due to nearby flaw coalesce. Both linear elastic and elastic-plastic fracture mechanics will be used to predict crack propagation and safe life under loading. Uncertainty quantification techniques will be explored to bound fracture properties for AM materials. Solutions will be compared to experimental data. Example problems will be presented in the paper. For surface flaws and embedded flaws of regular shapes such as keyhole in AM parts, commercial tools such as Nasgro software and in house damage tolerance code based on Murakami stress intensity factor are used to calculate crack growth. Nasgro and Paris crack growth equations are used respectively in each code. The in-house code based on Murakami stress intensity factors runs much faster, which is important for subsequent probabilistic damage tolerance analyses. Damage tolerance analysis an edge crack in AM aluminum bar is performed. The bar has a width of two inches and an initial crack length of 0.05 inches, subjected to tensile load transverse to the crack. Nasgro TC02 and in-house code analyses provided similar crack growth results. For irregular flaws due to lack of fusion, finite element models are being built to evaluate stress intensity factors numerically and updated periodically with crack growth. Near surface volumetric flaws will also be modelled by finite element method to capture the crack growth and rapid transition to large surface flaw behavior. When loading exceeds linear elastic fractur mechanics (LEFM) limit, the elastic-plastic fracture mechanics (EPFM) module of Nasgro software will be used to calculate safe life, and compared to LFEM prediction. Uncertainties in initial flaw size, aspect ratio and spatial distribution, and loading are being studied. Variabilities in crack growth curve of AM parts are bounded by probabilistic distribution. Monte Carlo simulation of in-house damage tolerance code is used to calculate probability of failure for a given number of cycles. Surrogate modelling will be used to speedup probabilistic damage tolerance analysis.