Session: DTM-03: Computational Methods in Design

Paper Number: 143509

143509 - Valuing Outliers: a Modeling Framework to Consider Non-Traditional Solutions From Non-Traditional Solvers 

While the engineering design research community has developed numerous models for early-stage design over the years, these models explore the design space corresponding to the dominant solution in the domain. These models work well for making predictions for designs by experts within an organization, but they cannot compare solutions that use alternative principles to solve the same problem. With the increasing popularity of open innovation and crowdsourcing within engineering design organizations, it is important to value the non-traditional solutions from non-traditional solvers in the early-stage design to make informed decisions about structuring the problem-solving mechanism employed by the organization. This paper attempts to create a new modeling framework to predict the design outcome by valuing non-traditional solutions alongside the dominant solution, thereby capturing the potential performance and novelty offered by the solutions from the non-traditional solvers. The model is demonstrated using a case study of a robotic gripper design. The results indicate the model can predict the best solution for the same problem across different principles. In addition, we find that the model is more important for non-traditional solvers than traditional solvers, as a larger fraction of their solutions would have been missed by models considering only the dominant solution.

Presenting Author: Athul Chakkithara Dharmarajan Purdue University

Presenting Author Biography: Athul is a PhD student in the School of Mechanical Engineering at Purdue University.

Authors:

Athul Chakkithara Dharmarajan Purdue University
Taylan Topcu Virginia Tech
Jitesh Panchal Purdue University
Zoe Szajnfarber The George Washington University