Search

Undergrad Research

home_outline/Current Students/Undergrad Research
College of Engineering COVID-19 Resources »
U-M’s COVID-19 Data Dashboard »

Get involved

By working with nationally recognized faculty and graduate students, Industrial Operations and Engineering undergrad students are able to get real-world research experience in fields like healthcare, aviation, transportation and more.

What is it like to be an undergraduate researcher?

IOE SURE/SROP Projects – 2021

The College of Engineering and the University of Michigan offer two programs SURE (Summer Undergraduate Research in Engineering) and SROP (Summer Research Opportunity Program) to provide undergraduate students with an opportunity to participate in summer research. Please find information below about these two exciting programs. Read more here.

IOE Project 1 (for 1 – 2 students)

Human-autonomy/robot interaction

Faculty Mentor(s): Xi Jessie Yang xijyang@umich.edu
Research Mode: Hybrid

We conduct research to investigate how humans interact with automated technologies/robots. Students can work on various dimensions of a project including designing computer games, coding, conducting human-subject experiments and data analysis. For students interested in game design and coding, a strong background in EECS is expected (java, c++, python etc.). For students interested in experiments and data analysis, some experience with statistics is preferred. To check out the specific projects a student can work on, please visit http://icrl.engin.umich.edu.

IOE Project 2 (for 1 – 2 students)

Applications of Industrial Engineering and Data Analytics to Medical Decision Making and Healthcare

Faculty Mentor(s): Mariel Lavieri lavieri@umich.edu
Research Mode: Remote

My research team addresses a variety of questions related to the development and application of Industrial Engineering methodology to support medical decision making and other healthcare decisions. Using large datasets, some of the sample questions investigated by our team include: when to monitor and treat patients with chronic conditions? How to plan the long-term supply and demand for transplant organs? How to prevent hospital readmissions? These questions are addressed both from a patient and from a system’s perspective.

The student(s) involved in this research will support model development, validation and implementation potentially including (but not limited to):

  • Literature searches
  • Data gathering/data mining
  • Performance of statistical analysis
  • Development and coding of algorithms
  • Development of simulation models
  • Validation of findings
  • Development of user friendly tools to implement findings (through possibly web based applications)

No prior medical/healthcare experience is necessary. Students involved will work actively with the research team and participate in our lab meetings both in Engineering and at the Hospital. Students with good communication and programming skills and some statistical/machine learning background (or willingness to acquire those skills) will be preferred.

IOE Project 3 (for 1 – 2 students)

Scalable Optimization Methods With Applications in Machine Learning

Faculty Mentor(s): Salar Fattahi fattahi@umich.edu
Research Mode: Online, Remote, or Hybrid

In this project, our aim is to develop and implement efficient computational methods for data-driven optimization problems with a wide range of applications in machine learning, including graphical model inference, robust/sparse PCA, and safe reinforcement learning. The students will have the opportunity to: (1) work on the theoretical aspects of these problems, (2) develop practical algorithms with certifiable guarantees, (3) implement their algorithms in Python/MATLAB/C++, and (4) test their developed algorithms on real datasets collected from brain networks, power systems, transportation networks, and financial markets. Students with background in optimization, machine learning, and programming (or the desire to acquire these skills) are preferred.

The admitted students are expected to:

  • Read and understand related literature
  • Collaborate with 1-2 PhD students
  • Develop efficient codes, and validate their correctness
  • Evaluate the performance of their developed codes on real-world data
  • Report their progress in weekly meetings

IOE Project 4 (for 1 – 2 students)

Approximation and Online Algorithms

Faculty Mentor(s): Viswanath Nagarajan viswa@umich.edu
Research Mode: Remote or Hybrid

This project will investigate efficient algorithms for approximately solving “hard” optimization problems. The specific problem(s) considered can be from a variety of applications, such as telecommunication, supply-chain management and scheduling.

Students will learn the following:

  • designing algorithms with mathematically rigorous performance guarantees.
  • implementing these algorithms in Python (or another programming language).
  • data collection.
  • testing algorithms computationally.
  • reading papers on related topics.

Collaboration with graduate students is also expected. Background in optimization, understanding mathematical proofs and programming is preferred.
 

IOE Project 5 (for 1- 2 students)

Fast, Adaptive and Scalable Stochastic Optimization Algorithms for Deep Learning

Faculty Mentor(s): Albert Berahas aberahas@umich.edu
Research Mode: Remote, Online

Deep learning has emerged as one of the most popular paradigms in machine learning due to its unprecedented successes in many domains such as computer vision, speech and image recognition, and machine translation. The goal of this project is to develop some of the next generation algorithms for training deep neural networks. Our algorithm(s) will be endowed with the following characteristics: fast, algorithms with sound theoretical guarantees and good practical performance; adaptive, algorithms with minimal dependence on hyper-parameters; and, scalable, algorithms able to solve problems with millions of variables. The student(s) involved in this research project will participate in all aspects of the development of the algorithms (algorithmic, theoretical, computational).

Students will be expected to:

  • Read, understand and present related literature.
  • Develop prototype codes for initial algorithmic testing.
  • Analyze (mathematically) the properties of the algorithms developed.
  • Develop efficient implementations of the algorithms developed.
  • Report progress in weekly meetings.
  • Collaborate with the research group.
  • Write a report of findings at the end of the program.

No prior research experience required. Elementary knowledge of Python required.
 

IOE Project 6 (for 1 – 4 students)

Applied Optimization Projects in Healthcare

Faculty Mentor(s): Amy Cohn amycohn@umich.edu
Research Mode: Hybrid

We will work on the application of modeling and operations research techniques to applied problems in healthcare. Prior projects have included work in transplant surgery, emergency medicine, and precision health. Students will have the opportunity to work with a wide variety of other students as well as experts from the application domains (e.g. physicians, nurses, clinical managers). Desirable skills and background include IOE310, programming skills, data analysis skills, and strong writing and interpersonal skills.
 

IOE Project 7 (for 1 student)

Supply Chain Design in the Face of Supply and Demand Uncertainty

Faculty Mentor(s): Mark S. Daskin msdaskin@umich.edu
Research Mode: Hybrid

We will be continuing work with a major automotive manufacturer on the design of their supply chain in the face of demand uncertainty, exchange-rate uncertainty, freight rate uncertainty, and supplier disruptions. Key questions include determining how many suppliers to have, where those suppliers should be across the globe, how much capacity each supplier should have, and how to deal with supply and demand disruptions after uncertainty is revealed. The project requires a fundamental understanding of probability and statistics, as well as optimization modeling.
 

IOE Project 8 (for 1 student)

Models for Higher Education Financing and Funding

Faculty Mentor(s): Mark S. Daskin msdaskin@umich.edu
Research Mode: Remote

Higher education was facing a significant crisis even before the beginning of the COVID-19 pandemic. Nationally, student debt amounted to about $1.6 trillion. The cost of tuition and fees has grown about four times as fast as has the consumer price index since 1982. Enrollment in higher education is down about 10 percent from its high in 2011. In most states, funding for higher education has not recovered from the last recession. Increasingly, many are questioning the value of higher education. Many private-for-profit institutions have closed in the last decade as have some private not-for-profit colleges and universities. COVID-19 has only exacerbated many of these phenomena. In this project, we try to understand these trends, estimate the costs of various policies including “free public tuition,” and “eliminating student debt.” We also try to develop predictive models that will assess the likelihood of an institution closing.
 

IOE Project 9 (for 1- 2 students)

Increasing stroke patient independence via personalized remote rehabilitation

Faculty Mentor(s): Bernard Martin martinbj@umich.edu
Research Mode: Lab and Remote

We conduct research in the development of rehabilitative methods for stroke patients that can be performed remotely, thereby increasing their independence. Our aim is to develop an app containing incentivized games that can be played to improve proprioceptive and motor abilities, which is geared to address the patient’s primary deficits. Students would work on the design and development of this gaming app. They will be provided with instructions as to what the games should include and dedicate their time to programing it. To accomplish this, a strong background in EECS is expected (java, c++, python etc.).

The student(s) involved in this research will support app development and implementation potentially including (but not limited to):

  • Literature searches
  • Development and coding of algorithms
  • Data collection and processing
  • Validation of findings

IOE Project 10 (for 1 student)

Estimating Drug Wastage at the University of Michigan Health System

Faculty Mentor(s): Mark S. Daskin msdaskin@umich.edu
Research Mode: Hybrid

Estimating the demand for drugs is a challenge for every healthcare system in the country. Many healthcare systems try to avoid shortages of drugs, particularly those drugs that are needed to save patient lives. Drugs also have expiration dates beyond which they should not be used and they must be disposed. The criticality of many drugs coupled with the limited time during which they can be used results in drug wastage. The goal of this project is to help Michigan Medicine estimate the cost of outdated drugs. The student will work closely with a PhD student in the IOE department and with faculty in both the IOE department and the Pharmacy department on this project. Strong computer skills and a good knowledge of basic statistics is required.
 

IOE Project 11 (for 2-3 students)

Quantifying Human Performance for Operational Environments

Faculty Mentor(s): Leia Stirling leias@umich.edu
Research Mode: Hybrid

In this interdisciplinary research group, we bring together methods from human factors, biomechanics, and robotics. We strive to understand the physical and cognitive interactions for goal-oriented human task performance and support operational decision making that relies on manual task performance. These goals may include reducing musculoskeletal injury risks, supporting telehealth, and improving technology usability. However, the term performance is not universally defined and requires learning about the desired task goals and the sub-tasks and motions the human will need to accomplish them.
 
There are two projects students may support.

  • Upper Extremity Exoskeletons for Industrial Applications. Exoskeletons are currently being evaluated for many different applications. In this project, the student will support the development and pilot testing of a study evaluating upper extremity exoskeletons designed to support overhead work.
  • Quantifying Tool Usage to Support NASA Astronaut Lunar Missions. As NASA plans to go back to the moon, decisions are being made on which tools to bring and how those tools should be used. In this project, the student will support the development and pilot testing of a study to evaluate lunar tools for geological sample collection.