Overall challenges in incorporating micro-mechanical models into materials design process

Abstract

Using materials in engineering design has historically been handled using the paradigm of selecting appropriate materials from the finite set of available material databases. Recent trends, however, have moved toward the tailoring of materials that meet the overall system performance requirements, based on a process called material design. An important building block of this process is micromechanical models that relate microstructure to proprieties. Unfortunately, these models remain short and include a lot of uncertainties from assumptions and idealizations, which, unavoidably, impacts material design strategy. In this work, candidate methods to deal with micromechanical models uncertainties and their drawbacks in material design are investigated. Robust design methods for quantifying uncertainty and managing or mitigating its impact on design performances are reviewed first. These methods include principles for classifying uncertainty, mathematical techniques for evaluating its level degree, and design methods for performing and generating design alternatives, that are relatively insensitive to sources of uncertainty and flexible for admitting design changes or variations. The last section of this paper addresses the limits of the existing approaches from material modelling perspective and identifies the research opportunities to overcome the impediment of incorporating micromechanical models in material design process.