You are here

Quantifying Uncertainty in the Mechanical Performance of Additively Manufactured Parts Due to Material and Process Variation


TECHNOLOGY AREA(S): Materials/Processes


OBJECTIVE: Quantify the effect of variations in process characteristics on the mechanical performance of additively manufactured (AM) parts, and develop a procedure for mitigating these effects within statistical bounds using Integrated Computational Materials Engineering (ICME) framework.

DESCRIPTION: The certification of additively manufactured parts requires that mechanical performance be quantified to a minimum of B-basis allowable (90% of the population values are expected to equal or exceed that strength value with 95% confidence). Additional considerations will be necessary based on the artifacts reliability and functionality, for example in the determination that the part is a single point of failure will elevate this to an A-basis allowable (at least 99% of the population values is expected to equal or exceed this tolerance bound with 95% confidence). To achieve this, the Navy requires quantitative uncertainty assessment methods to predict, with a known level of confidence, the microstructural and mechanical property outcomes for a specific material, machine, geometry, and post-processing combination involved in an AM system.

A material used in an AM process undergoes several complex, transient, and interacting physical phenomena, including: heat and mass transfer, material phase transformation, and free-surface fluid flow. These phenomena significantly affect the material property distribution of a built component. AM post-processing techniques, such as stress relief and hot isostatic pressing (HIP), further alter the distribution of material properties obtained during the deposition process. Therefore, approaches such as ICME are needed to link the time and length scales of the occurring physical phenomena. These multi-scale simulations should predict: the interdependencies at play among deposition process; the resulting material micro-structure; local mechanical properties; the overall component performance; and the effects of post-processing.

The challenge inherent to AM processes is the mechanical property distributions within a specific part which are functions of stochastic variables. Knowing the exact machine and material state at any point in time has inherent uncertainties in the form of aleatoric and epistemic uncertainties. The main sources of aleatory uncertainty in AM systems include material characteristics (e.g. chemical composition, powder size distribution, roundness) and process parameters (e.g. laser scan speed, power density, and delay time). There are also several sources of epistemic uncertainty present in AM such as powder local compaction density, friction between powder particles and so on. To effectively model this highly complex and random process, powerful stochastic modeling techniques such as in reference [1] are needed, connecting material and processing characteristics to microstructure distribution, mechanical property distribution, and mechanical performance.

The challenge is to determine how multiple sources of uncertainties are propagated in a model developed specifically for an AM process, such as in reference [2], and then how to quantify the uncertainty of the resulting material properties and microstructure to predict desired performance in probabilistic terms. Keeping this challenge in mind, the topic requires: a comprehensive approach [3] to quantify the uncertainties of material and process model parameters; recommendations on minimizing both material and process uncertainties in production; and suggestions for acceptance metrics/criteria and tolerances for decision making.

One approach could be the use of physics based models or ICME tools to run simulations of the AM process to narrow down the uncertainty.

One approach could be the use of physics based models or ICME tools to run simulations of the AM process to narrow down the uncertainty.

PHASE II: Further develop and finalize the concept, processing methodology and/or tool from Phase I for metallic materials relevant to naval aviation. Design and perform experiments to validate the approach and to quantify uncertainty in standard test methods for determining material and process characteristics. Develop an uncertainty analysis method to assess the impact of parameter/model uncertainties on the output of metallic AM parts certification approach.

PHASE III DUAL USE APPLICATIONS: Deliver a capability to provide rapid uncertainty quantification for the mechanical performance of a broad range of additively manufactured metallic parts. These new approaches can be used to accelerate the FAA certification process as well as the NAVAIR process. Fast uncertainty quantification will promote a wider acceptance of AM technology within both the military and commercial sector.


    • Stefanou, G., 2009, "The stochastic finite element method: past, present and future," Computer Methods in Applied Mechanics and Engineering, 198(9), pp. 1031-1051.


    • Pal, D., Patil, N., Zeng, K., and Stucker, B., 2014, "An Integrated Approach to Additive Manufacturing Simulations Using Physics Based, Coupled Multiscale Process Modeling," Journal of Manufacturing Science and Engineering, 136(6), p. 061022.


  • Roy, C.J., and Oberkampf, W.L., 2010, "A Complete Framework for Verification, Validation, and Uncertainty Quantification in Scientific Computing" Proceedings of the 48th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition, Orlando, Florida.

KEYWORDS: Additive Manufacturing; Modeling; Metallic; Microstructure; Materials Processing; Quantification

  • TPOC-1: 301-342-5169
  • TPOC-2: 301-342-9389

Questions may also be submitted through DoD SBIR/STTR SITIS website.

US Flag An Official Website of the United States Government