Characterizing Implicit Cognitive Process in Engineering Design Across Contexts using Machine Learning Methods

open_access
Open Access
Author:
Malviya, Manoj
Graduate Program:
Mechanical Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
June 29, 2020
Committee Members:
Catherine G P Berdanier, Thesis Advisor/Co-Advisor
Christopher Carson Mccomb, Committee Member
Daniel Connell Haworth, Program Head/Chair
Keywords:
Engineering
Design
Machine Learning
Education
Design Thinking
Abstract:
Researchers have proposed many psychological theories of design, composition, problem solving and reasoning in literature, highlighting the role of implicit cognitive processes. Particularly in an ill-defined or complex problem setting, there is a scarcity of literature resulting in limited impact of the field of creative problem solving. This research employs stochastic machine learning methods across three aspects of engineering problem solving and design that typically are not studied together: low-fidelity engineering design settings (in this study, studying conceptual problem-solving design thinking using the Delta Design challenge), high-fidelity engineering design settings (in this study, in an Additive Manufacturing and Design Context) and in non-traditional “design” contexts (in this study, investigating design principles as engineers write research proposals.) The purpose of this research is to characterize the “design” processes in each of these contexts, working toward a unified theory of engineering implicit cognition during activities requiring creation or design. The population of interest for each of these studies is graduate engineering students, which is a unique population to typical engineering education or design thinking literature.