MAE 531: Engineering Design Optimization Fall 2023

Mechanical and Aerospace Engineering North Carolina State University

MAE 531: Engineering Design Optimization

Fall 2023

SYLLABUS AND SEMESTER PLAN

Course summary
• This class will focus on strengths and weaknesses of various optimization methods and discuss best-
practices around problem formulation.
• Graded material will consist of homework and three coding and reflection assignments.
• Optimization methods will be implemented using hand-calculations, Excel, and Matlab. Basic coding
experience is required.

Course instructor, times, and location

Instructor: Dr. Scott Ferguson

[email protected] ([email protected] can also be used as they go to the same account)

Lecture: Mondays and Wednesdays from 3:00 pm – 4:15 pm

2207 Engineering Building 3

Office hours: Mondays and Wednesdays from 12:00 pm – 1:00 pm (or by scheduled appointment)
3244 Engineering Building 3
(As needed, virtual office hours can also be arranged)

Course description

Optimization involves finding the “best” solution according to specified criteria. In the context of engineering
design, the “best” solution may refer to a minimum cost or weight, maximum quality or efficiency, or some other
performance index pertaining to a disciplinary objective. However, determining the optimal design involves more
than just the minimization or maximization of an objective function. Designers must also identify the design
variables that represent the physical form of the system and the constraints that represent limitations on the design
space. Typically, the problems of interest in engineering are of a nonlinear nature, in that the objective functions
and constraints considered are nonlinear.

This course introduces traditional nonlinear optimization methods that can be used to solve a wide variety of
engineering design problems across all engineering disciplines. Additionally, students will study the tradeoffs
associated with the design of complex engineering systems. By the end of the semester, students will have gained
the ability to create a formal optimization problem, have basic knowledge of numerical optimization algorithms,
and will have sufficient understanding of the strengths and weaknesses of these approaches.

Note: Some of the homework and both implementation projects involve programming. A basic programming
background is required (for loops, if-else statements, while, etc.). We will cover some Matlab programming in
lecture, and some assignments will require you to program in Matlab. If you have not written code in a while, it is
strongly recommended that you read a Matlab tutorial or primer (I will also post some supplementary material).

Teaching philosophy for this course

“There is an intimate relationship between the mathematical model that describes a design and the solution
methods that optimize it. A basic premise from the start was that a good model can make optimization almost
trivial, whereas a bad one can make correct optimization difficult or impossible. Software tools today provide
capabilities for intricate analysis of many difficult performance aspects of a system. These analysis models, often
referred to as simulations, can be coupled with numerical optimization software to generate better designs
iteratively. … The success of such attempts depends strongly on how well the design problem has been formulated
for an optimization study, and on how familiar the designer is with the working and pitfalls of mathematical
optimization techniques.”
Panos Y. Papalambros, 2017, Principles of Optimal Design: Modeling and Computation, Third Edition

Updated: August 17, 2023 1

MAE 531: Engineering Design Optimization Fall 2023
Mechanical and Aerospace Engineering North Carolina State University

“The lasting measure of good teaching is what the individual student learns and carries away.”
Barbara Harrell Carson, 1996, Thirty Years of Stories

I will always be available during office hours and by appointment (unless otherwise specified). I will also try my
best to answer all email questions and Piazza posts within one business day. If you experience problems with the
material, please contact me for help. Finally, constructive course feedback is always encouraged.

Course topics

• Introduction to optimization – design variables, constraints, objective functions, penalty functions,

development of formalized optimization problem statements
• Modeling engineering design problems for optimization
• Using graphical techniques to identify an optimal problem
• Techniques for solving single variable optimization problems
• Techniques for solving unconstrained and constrained multi-variable problems
• Metamodeling and response surface generation
• Heuristic-based optimization methods
• Mathematical foundations of multiobjective and multidisciplinary design optimization

Skills/understanding

At the end of this class, you will have the foundation needed to:
• Formulate optimization problems in standard form
• Assess the optimality of a proposed solution
• Interpret the output of an optimization algorithm
• Assess the sensitivity of an optimal solution
• Compare the advantages and disadvantages of different optimization techniques
• Choose the best optimization technique for meeting your available resources
• Create metamodels using sampled problem data
• Solve multiobjective and multidisciplinary optimization problems
• Construct computer programs for determining the optimal solution for unconstrained and constrained
nonlinear optimization problems of multiple variables

Course websites

Moodle: http://wolfware.ncsu.edu
Panopto: https://ncsu.hosted.panopto.com/Panopto/Pages/Sessions/List.aspx#folderID=88f32844-c198-4016-85e3-b056010097e9
Piazza: https://piazza.com/ncsu/fall2023/mae531/home
Message board where students can ask questions and receive guidance from others and the professor

Text (not required, but useful)

Arora, J., Introduction to Optimum Design, 4th edition, Elsevier, ISBN 978-0-12-800806-5

All students will have free access to the digital book until September 6th. Students wanting to get the All-In
materials at the lowest price must place their order for the books before September 6th.

https://shop.ncsu.edu/adoption-search-results?ccid=65491&itemid=81411

Grading

Homework 40%
Implementation and reflection projects (3) 60% (20% each)

Updated: August 17, 2023 2

MAE 531: Engineering Design Optimization Fall 2023
Mechanical and Aerospace Engineering North Carolina State University

Homework:
Students will submit individual homework. The objective of homework is gaining experience with the concepts/
algorithms discussed in class. Homework will involve written work, discussion, some coding, and some graphing.

Implementation and reflection projects:

There will be three (3) implementation projects. Students will submit individual projects. These projects will
involve the computer-based implementation of optimization techniques covered in class. We will use problems
that have a larger scale than what we cover in the homework. The focus of these projects will be assessing the
effectiveness of the algorithms and comparing their performance in arriving at the optimal solution. In these
projects, getting the algorithm to work is only part of the assignment. You will also we asked to provide written
reflections that demonstrate what you have learned about problem formulation, algorithm effectiveness, and
solution quality.

Submission of homework and projects:

All homework and projects will be submitted via Moodle. A submission link will be created for each assignment.

For homework, students are allowed one 24-hour grace period on one assignment. You do not need to contact me
about this. Late assignments submitted without communication, and not used as part of the one grace period, will
not be accepted without justification. Students are recommended to see the University Attendance Policy for
reasons that late homework may be accepted.

For projects, no grace periods are provided. If you need extra time on a project, please contact me so that we can
discuss an extension. Communication about illness, travel, challenges, etc. is fundamentally important. Please
reach out early so we can develop a plan.

Miscellaneous

Health and participation in class:

Due to the COVID-19 pandemic, public health measures continue to be implemented across campus. Students
should stay current with these practices and expectations through the Protect the Pack website
(https://www.ncsu.edu/coronavirus/).
• If you test positive for COVID-19, or are told by a healthcare provider that you are presumed positive for
the virus, you should not attend any hybrid or face-to-face (F2F) classes and work with your instructor
on any adjustments necessary; also follow other university guidelines, including self-reporting
(Coronavirus Self Reporting): Self-reporting is not only to help provide support to you, but also to assist
in contact tracing for containing the spread of the virus.
• If you feel unwell, even if you have not been knowingly exposed to COVID-19, please do not come to a
F2F class or activity.
• If you are in quarantine, have been notified that you may have been exposed to COVID-19, or have a
personal or family situation related to COVID-19 that prevents you from attending this course in person
(or synchronously), please connect with your instructor to make alternative plans, as necessary.
• If you need to make a request for an academic consideration related to COVID-19, such as a discussion
about possible options for remote learning, please talk with your instructor.
• All members of the NC State academic community are expected to follow all university policies and
guidelines, including the Personal Safety Rule and community standards, for the use of face
coverings. Face coverings are required in instructional spaces. Face coverings should be worn to cover
the nose and mouth and be close fitting to the face with minimal gaps on the sides.

Course attendance:
NC State attendance policies can be found at: https://policies.ncsu.edu/regulation/reg-02-20-03-attendance-
regulations/ . Please refer to this course’s attendance, absence, and deadline policies for additional details. If you
are quarantined or otherwise need to miss class because you have been advised that you may have been exposed
to COVID-19, you should not be penalized regarding attendance or class participation. However, you will be
expected to develop a plan to keep up with your coursework during any such absences. If you become ill with
COVID-19, you should follow the steps outlined in the health and participating section above. COVID 19-related
absences will be considered excused; documentation need only involve communication with your instructor.

Updated: August 17, 2023 3

MAE 531: Engineering Design Optimization Fall 2023
Mechanical and Aerospace Engineering North Carolina State University

Technology requirements:
This course may require particular technologies to complete coursework. Be sure to review the syllabus for these
expectations and see go.ncsu.edu/syllabus-tech-requirements to find out more about technical requirements for
your course. If you need access to additional technological support, please contact the Libraries’ Technology
Lending Service: https://www.lib.ncsu.edu/devices.

Academic dishonesty:
Cheating of any kind (copying, plagiarism, etc.) will not be tolerated and will result in an F for the course. Please
review the following website (http://www.ncsu.edu/stud_affairs/osc/AIpage/acaintegrity.html) for further
information on academic integrity at NCSU. In addition, reasonable accommodations will be made for students
with verifiable disabilities. Students must register with the Disability Services Office.

Supporting fellow students in distress:

As members of the NC State Wolfpack community, we each share a personal responsibility to express concern for
one another and to ensure that this classroom and the campus as a whole remains a safe environment for learning.
Occasionally, you may come across a fellow classmate whose personal behavior concerns or worries you. When
this is the case, I would encourage you to report this behavior to the NC State Students of Concern
website:http://studentsofconcern.ncsu.edu/. Although you can report anonymously, it is preferred that you share
your contact information so they can follow-up with you personally.

Updated: August 17, 2023 4

MAE 531: Engineering Design Optimization Fall 2023
Mechanical and Aerospace Engineering North Carolina State University

Course outline (dates, topics, and assignments all subject to change)

Week Topic Assignment

Introduction to optimization
1 Terminology
Problem preparation and formalized optimization problem statements
Development of optimality conditions
2 Existence and uniqueness of solutions; Necessary and sufficient conditions
Finding and assessing the optimum using graphical approaches
Optimizing functions of a single variable
3 Bounding phase method HW 1 (due 9/6)
Point-estimation techniques
Optimizing functions of a single variable
Region elimination techniques
Unconstrained functions of multiple variables
4
Iterative solution technique
Zero-order methods
Gradient estimation techniques
Unconstrained functions of multiple variables
5 Gradient-based methods HW 2 (due 9/18)
Second order methods
6 Basics of linear programming HW 3 (due 9/27)
7 Linear programming extensions to nonlinear optimization problems
Constrained nonlinear optimization (SUMT methods)
Interior penalty function
8 HW 4 (due 10/11)
Exterior penalty function
Augmented Lagrange Multiplier (ALM)
Constrained nonlinear optimization Implementation and
9 Generalized Reduced Gradient reflection project 1
SQP (due 10/18)
Procedures for building a model
10 HW 5 (due 10/23)
Metamodeling and response surface generation
11 Advanced metamodeling and response surface generation HW 6 (due 11/1)
Heuristic optimization techniques Implementation and
12-13 Simulated annealing reflection project 2
Particle swarm optimization (due 11/13)
Heuristic optimization techniques
13-14
Genetic algorithms
Optimization problem framing
Multiobjective optimization Implementation and
15-16 Pareto sets and tradeoffs reflection project 3
Basic multiobjective formulations (weighted sum, compromise programming) (due 11/29)
NSGA-II

Updated: August 17, 2023 5