Are you sure about that? Introducing Uncertainty in Undergraduate Engineering

Sophia V. Yates; Christopher H. Conley; Aaron J. Rubin

Download Paper | Permalink

Conference: 2021 ASEE Virtual Annual Conference Content Access
Location: Virtual Conference
Publication Date: July 26, 2021
Start Date: July 26, 2021
End Date: July 19, 2022
Conference Session: Design Related
Tagged Division: Mechanical Engineering
Page Count: 17
DOI: 10.18260/1-2--36698
Permanent URL: https://strategy.asee.org/36698
Download Count: 293

Request a correction

Paper Authors

biography

Sophia V. Yates Smith College

visit author page

Sophia V. Yates is a junior studying engineering sciences in the Picker Engineering Program at Smith College. She is interested in structural engineering and engineering education. Yates is planning on pursuing a graduate degree in STEM education after her time at Smith College.

visit author page

biography

Christopher H. Conley Smith College

visit author page

Chris Conley is currently a research affiliate with the Picker Engineering Program at Smith College. He has served on the faculties of four institutions over the last three decades. He has also worked as a researcher at four federal laboratories. His research interests include modeling and simulation, both physical and numerical, with special interest in the response of structures to extreme loads. Helping others learn to apply engineering mechanics to better understand their world is a passion.

visit author page

biography

Aaron J. Rubin Smith College

visit author page

Aaron J. Rubin is a lecturer at Smith College where he teaches Junior and Senior level undergraduate engineering courses including Finite Element Modeling and Senior Design Clinic.

visit author page

Download Paper | Permalink

Abstract

George Box famously stated “all models are wrong, but some are useful” in a 1978 workshop on Robustness in Statistics. Building on that notion, we explore how undergraduate engineering students may begin to learn to differentiate the "useful" models and the limits of their usefulness. In the last forty years, the field of engineering has become increasingly reliant on computer-based modeling. Usage of numerical modeling and simulation in the undergraduate engineering curriculum has also expanded. The increase in usage of this powerful tool is not without consequences. Engineers often neglect the potential impact of the simplifications that go into creating a model and thus neglect the uncertainties inherent in simulation results. This high faith in models can lead to incorrect conclusions and potentially catastrophic consequences.

In this paper, the authors discuss the necessity and utility of learning about uncertainty in conjunction with modeling and simulation at the undergraduate level. They provide content to scaffold a student’s intuition about uncertainty while also enhancing the student’s understanding of modeling in multiple disciplines. The authors use a case study that compares a finite element analysis (FEA) simulation to data from physical experiments as a foundation for learning about modeling and simulation. Comparing physical experiments and results of numerical simulations, the students will investigate how the results were affected by assumptions. Quantifying the impact of their assumptions through analysis of uncertainty will accompany the validation of their simulations.

Based on the literature reviewed, the current treatment of uncertainty in numerical modeling follows the general form of the quantification of uncertainty in physical modeling. Therefore, the lesson content on quantification of uncertainty reviews what is commonly covered when discussing physical experiments, and demonstrates how that can be extended into the numerical modeling realm. Both Type A and Type B evaluations of uncertainty are covered. Incorporated throughout the educational tools presented in this paper is a vocabulary necessary to discuss computer-based numerical simulations.

Citation
Format

Yates, S. V., & Conley, C. H., & Rubin, A. J. (2021, July), Are you sure about that? Introducing Uncertainty in Undergraduate Engineering Paper presented at 2021 ASEE Virtual Annual Conference Content Access, Virtual Conference. 10.18260/1-2--36698

TY - CPAPER
AB - George Box famously stated “all models are wrong, but some are useful” in a 1978 workshop on Robustness in Statistics. Building on that notion, we explore how undergraduate engineering students may begin to learn to differentiate the &quot;useful&quot; models and the limits of their usefulness. In the last forty years, the field of engineering has become increasingly reliant on computer-based modeling. Usage of numerical modeling and simulation in the undergraduate engineering curriculum has also expanded. The increase in usage of this powerful tool is not without consequences. Engineers often neglect the potential impact of the simplifications that go into creating a model and thus neglect the uncertainties inherent in simulation results. This high faith in models can lead to incorrect conclusions and potentially catastrophic consequences.

In this paper, the authors discuss the necessity and utility of learning about uncertainty in conjunction with modeling and simulation at the undergraduate level. They provide content to scaffold a student’s intuition about uncertainty while also enhancing the student’s understanding of modeling in multiple disciplines. The authors use a case study that compares a finite element analysis (FEA) simulation to data from physical experiments as a foundation for learning about modeling and simulation. Comparing physical experiments and results of numerical simulations, the students will investigate how the results were affected by assumptions. Quantifying the impact of their assumptions through analysis of uncertainty will accompany the validation of their simulations.

Based on the literature reviewed, the current treatment of uncertainty in numerical modeling follows the general form of the quantification of uncertainty in physical modeling. Therefore, the lesson content on quantification of uncertainty reviews what is commonly covered when discussing physical experiments, and demonstrates how that can be extended into the numerical modeling realm. Both Type A and Type B evaluations of uncertainty are covered. Incorporated throughout the educational tools presented in this paper is a vocabulary necessary to discuss computer-based numerical simulations.

AU - Sophia V. Yates
AU - Christopher H. Conley
AU - Aaron J. Rubin
CY - Virtual Conference
DA - 2021/07/26
PB - ASEE Conferences
TI - Are you sure about that? Introducing Uncertainty in Undergraduate Engineering
UR - https://strategy.asee.org/36698
DO - 10.18260/1-2--36698
ER -