The proposed research targets advancing the state of knowledge on rational design decisions in sustainable product design in order to enable a better understanding of the consequences of those decisions from an overall design perspective. The proposed research method will advance the current available capabilities for product and process design by integrating LCA into decision making for engineering design. The work intends to result in a novel decision method, and related decision support tools, for sustainable design. The methods and tools to be developed reflect the standards established in NISTs sustainability portal, and complement and enhance the capabilities of PLM tools.
The outcomes of the proposed work have the potential to facilitate the design of more sustainable products that could significantly improve both the environment and the economy on a global scale. The work is supported by the Industry Advisory Board as well as individual industry members of the center and has the potential to extend the centers portfolio. The PIs plan to introduce the developed research methods in the graduate courses at UMass Amherst and the University at Buffalo SUNY as well as develop coordinated prototype educational tools to benefit the centers collaborators and industry partners.
A novel semantic framework was created to model the information of the domains necessary for the sustainable design of products. This unique approach considers both compliance with the applicable standards and also objectives compatible with triple bottom line benefits to the economy, environment, and stakeholders, in terms of performance delivered. Since the applicable standards and criteria are contained within the same information model in real time, the standards may be adopted more easily early on while the design may also be influenced more toward the triple bottom line objectives. Furthermore, the design intent is captured and revealed transparently to all design participants dynamically. The case studied showed that sustainable design may be considered earlier in a design process in such cases where the optimal design for sustainability depends upon material and geometry input variables exclusively. Furthermore, the case examined shows that some consistencies can be revealed between applicable regulations modeled by standards and environmental impacts determined by LCA. The process enabled by the new framework was shown to allow parallel inspection of information related to standards and design alternative selection. A novel modeling and simulation method was developed that focused on material selection for the most significant effect on sustainability objectives in the early design of product components. Material selection is not conducive to more efficient and robust surrogate model construction due to the inflexible discrete locations of material related data points and the dimensionality of the data. To address this, the novel technique both streamlines the Life Cycle Assessment model construction for viable material alternatives and simplifies model dimensionality by the consolidation of factors. This development enabled the construction of robust surrogate models of environmental objectives in a rigorous representation with other traditional design objectives. A novel feasible approximation approach was conceived and developed for model sampling. This approach addressed the unique challenges posed by rigid data locations of material parameters. Robust results can be achieved by use of the new adapted Latin Hypercube approach at the first of two sampling stages. Such an approach enables optimal concept identification within an entire design space beyond the original data set of known design alternatives. Two examples illustrated the potential for reasonable robustness at the early design stages. A novel decision making method was also developed that considers designer preferences over multiple attributes and at multiple hierarchical levels. The proposed approach simplifies the handoff of data between LCA and HEIM, utilizing H-HEIM to present the Life Cycle Assessment data to the decision maker in a way that is easy to understand and allows them to make trade-offs between attributes at varying levels. It also provides the decision maker with increased information, by acknowledging that there is uncertainty in the weights assigned to various attributes and scaling the weights by the level of confidence in the accuracy of each local weight. These works combined to publish five conference papers and two journal papers.
