HOME    »    PROGRAMS/ACTIVITIES    »    
IMA/MCIM Industrial Problems Seminar

In collaboration with the University of Minnesota’s School of Mathematics, the Industrial Problems Seminars are a forum for industrial researchers to present their work to an audience of IMA postdocs, visitors, and graduate students, offering a first-hand glimpse into industrial research. The seminar series is often useful for initiating contact with industrial scientists. The IMA’s seminar series is the oldest and longest running seminar series in industrial mathematics.

This year's seminars are organized by Gilad Lerman, School of Mathematics, University of Minnesota.

Add seminars to your Google Calendar for reminders.


  • SIAM Internship Panel


    September 17, 2015 4:40 PM - 6:00 PM
    Kolthoff 137 [Map]

    Abstract

    The event will have graduate students speaking about their experiences with their summer internships and professors speaking about their experience facilitating internships. Following their talks, we will have a question and answer session. Food and beverages will be provided at the event. The participants of the Graduate Student Internship Panel will be: Prof. Gilad Lerman, Prof. Fadil Santosa, Brittany Baker (interned at NSA), Tyler Maunu (interned at Amazon & Schlumberger), Jeffrey Moulton (interned at Allstate & Air Force), Yu Zhou (interned at Seattle Genetics). The event is organized by the SIAM Student Chapter at the University of Minnesota.
  • Effective Strategies for Your Job and Internship Search

    Whitney Moore, University of Minnesota, Twin Cities
    September 25, 2015 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    The best advice for internship and job searching is to begin your search early! The process of looking for an opportunity is truly an individualized experience and it can take time to find and engage with employers who have opportunities that excite you the most. Developing and utilizing a network and establishing what is most important to you when it comes to industries, companies, and types of positions will go a long way in helping you have a strategic and purposeful job or internship search. This workshop will discuss strategies for building your network and approaches to help make your job or internship more effective.

    Whitney has been a Career Counselor in the CSE Career Center at the University of Minnestota since the fall of 2011. Her passions lie in helping students discover their skills and interests, understand how to navigate the career development and job search process, and connecting them to resources and opportunities that will help them be successful. In addition to working in the CSE Career Center, Whitney is President-Elect for the Minnesota College and University Career Services Association (MCUCSA) and Co-Chairs the MCUCSA Communication Committee. She has also served as Chair-Elect and Co-Chair of the Government and Nonprofit Career Fair Committee from 2012-2013 and been in two board positions for the Minnesota College Personnel Association. She received her B.A. in Communication Studies from Gustavus Adolphus College and M.S. in Counseling and Student Personnel from Minnesota State University, Mankato. Away from campus, Whitney enjoys all things Minnesota, including the North Shore, state parks, anything Swedish, and spending time running, biking, hiking, and fishing outdoors with her husband and rescue dog, Charli.

    Talk Slides
    Interviewing Guide
    Internship Job Search Guide
    Graduate Resume CV Guide
  • Manufacturing at Internet Speeds: CAD/CAM Challenges in Rapid Manufacturing

    Stefan Atev, Proto Labs, Inc.
    October 9, 2015 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    Powerful computer-aided design and manufacturing (CAD/CAM) software systems have been available for decades, yet the process of getting physical parts manufactured from a 3D CAD model is surprisingly manual and unchanged from the pre-Internet era. This talk will highlight Proto Labs’ unique perspective on digital manufacturing, explain why Proto Labs has a CAD/CAM software group and how the problems it tries to address differ from those of traditional CAD/CAM software companies. The talk will introduce the audience to some open problems such as searching for similar geometric models under representational and shape invariants.

    Stefan Atev is a Principal Software Developer for CAD/CAM at Proto Labs, Inc. Most of his two-year tenure at the company has been spent at the ProtoWorks R&D division, where he is working on problems related to CNC machining, mold design and production automation. He holds a PhD in Computer Science from the University of Minnesota (2011). Prior to Proto Labs, Stefan was a Senior Algorithm Scientist in the Medical Imaging group of Vital Images, Inc.
  • Mathematicians in Financial Services

    Mrinal Raghupathi, USAA Asset Management Company
    October 16, 2015 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    USAA is a large financial services company that provides banking, insurance and investment services and products to the military and their families. In this talk I will give an overview of the kinds of jobs available to mathematicians in financial services. The talk will be focused on some of the more practical issues faced in getting a job and navigating a non-academic environment. A brief list of topics:
    • How to get a job and be more marketable in financial services with a Math Ph.D.
    • What options exist in financial services (including the two roles I've had).
    • Skills that you need on a day-to-day basis.
    • Projects that you might work on (examples) and some of the practical hurdles to solving problems that you face in industry but not in academics.

    Mrinal Raghupathi is a lead quantitative analyst at USAA Asset Management Company. Prior to his career in financial services, Mrinal was an assistant professor in the department of mathematics at the US Naval Academy and a postdoc at Vanderbilt University. In his current role Mrinal analyzes risk for the USAA investment portfolios and researches quantitative investment methods. In his previous role he worked in model risk management. He received his Ph.D. in Mathematics from the University of Houston in 2008.
  • Mathematics at a 'Materials Company': How 3M Creates Advanced Technologies

    Nitsan Ben-Gal, 3M
    October 30, 2015 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    3M is a global innovation company developing products and solutions for the Industrial, Health Care, Safety, Graphics, Electronics, Energy, and Consumer sectors. While famous for its materials and processing capabilities, 3M’s Software, Electronic & Mechanical Systems Laboratory (SEMS) works with all areas of the company to develop innovative products and solutions utilizing Mathematical, Computer Science, and Engineering expertise.

    In this talk I will give an overview of some technologies in development or recently deployed within the Advanced Technologies Group that utilize its mathematical and computational expertise, and how these interplay with 3M’s established businesses. I will also discuss my own experience as a mathematician both on the job market and at 3M.

    Bio-sketch: Nitsan Ben-Gal is a Data Scientist in the 3M Corporate Research Laboratory, where she works on problems involving machine learning, predictive analytics, and computer vision. Prior to joining 3M, she completed a postdoc at the Institute for Mathematics and its Applications and the Weizmann Institute of Science. She holds a Ph.D. in Applied Mathematics from Brown University (2010).
  • Structured Data and Learning: Data Science that Embraces Complex, Heterogeneous, Relational Data

    Robert Nendorf, Allstate
    November 6, 2015 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    Invited by the SIAM Student Chapter at the University of Minnesota.

    Many businesses have heterogeneous data that models complex entities with rich relationships between them. This is especially true of insurance companies that must simultaneously set prices, underwrite policies, settle claims, manage agency operations, and deliver a great customer experience over time. Cutting edge technologies like graph databases and disciplines like structured relational learning now allow us to store data, develop ETL processes, and build analytics frameworks that leverage these rich relationships. We will discuss examples of graph querying and structured prediction in an applied context.

    Robert Nendorf is a data scientist in the Quantitative Research and Analytics department at Allstate. He is responsible for researching and prototyping data analytics frameworks to help the business make better decisions. This has included building social network analysis tools, predictive models, and big data reporting tools for both claims and pricing partners. Before that he was a mathematician at Northwestern University where he received his PhD in algebraic topology.
  • Magnetic Refrigeration From Laboratory to Products

    Steve Russek, Astronautics Corporation of America
    December 4, 2015 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    An overview of Astronautics Magnetocaloric Heat Pump research and development program will be presented. Magnetic Refrigeration is an emerging heat pumping technology with applications in refrigeration and air conditioning for residential and commercial settings. Magnetic Refrigeration technology is based upon the magnetocaloric effect in which a magnetic material undergoes a temperature change when it experiences a change in an applied magnetic field. By suitably integrating a magnetocaloric material into a process cycle using water-based heat transfer fluids a Magnetocaloric Heat Pump can be realized. When commercialized Magnetocaloric Heat Pumps will provide a Green, more efficient and lower noise alternative to conventional vapor compression based heat pumping technology.

    Dr. Steve Russek is the Director of Astronautics Technology Center based in Madison Wisconsin. Astronautics Corporation of America is a privately-held aerospace and defense contractor headquartered in Milwaukee Wisconsin. Prior to joining Astronautics Steve was an Engineering Development and Commercial Development Manager at Air Products and Chemicals. Steve has a BS in Chemical Engineering from Northwestern University and a PhD in Chemical Engineering from the University of California - Berkeley.
  • Towards Lights Out Electronics Manufacturing

    Jeff McAlvay, Tempo Automation
    December 11, 2015 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    As software plays an increasing role in the physical world (e.g., internet of things, wearables, autonomous cars), the demand for electronics has never been greater. Unfortunately, the experience of developing electronics today is dramatically worse than that of developing software. Instead of the speed and convenience of software iteration, electrical engineers are faced with a slow and tedious prototyping process. This is due to the labor-intensive nature of electronics manufacturing. Our goal is to infuse the production of electronics with software, thereby delivering a software-like development experience to electrical engineers. In this talk, we discuss various puzzles that must be resolved to execute on this vision, ranging from feature extraction from electronics design data, design for manufacturing analyses, as well as factory scheduling and coordination.
  • Interviewing Tips for Industry

    Whitney Moore, University of Minnesota, Twin Cities
    January 22, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    Whether you are applying for an internship or full-time job, the interview is one of the most important parts of the application process. This workshop will outline the different types of interviews and questions you can expect, how to best prepare and succeed in your interview, and provide you with suggested strategies you can use to answer some of the most commonly asked interview questions.

    Whitney has been a Career Counselor in the CSE Career Center at the University of Minnestota since the fall of 2011. Her passions lie in helping students discover their skills and interests, understand how to navigate the career development and job search process, and connecting them to resources and opportunities that will help them be successful. In addition to working in the CSE Career Center, Whitney is President-Elect for the Minnesota College and University Career Services Association (MCUCSA) and Co-Chairs the MCUCSA Communication Committee. She has also served as Chair-Elect and Co-Chair of the Government and Nonprofit Career Fair Committee from 2012-2013 and been in two board positions for the Minnesota College Personnel Association. She received her B.A. in Communication Studies from Gustavus Adolphus College and M.S. in Counseling and Student Personnel from Minnesota State University, Mankato. Away from campus, Whitney enjoys all things Minnesota, including the North Shore, state parks, anything Swedish, and spending time running, biking, hiking, and fishing outdoors with her husband and rescue dog, Charli.
  • Advanced Healthcare Research and Analytics: Catching fraud, Removing Waste, Finding Providers, and Happy Consumers

    Jason Haupt, UnitedHealth Group
    February 5, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    UnitedHealthcare's Advanced Research and Analytics (ARA) teams solve business problems across all internal lines of business. During this presentation we will discuss 3-4 high impact business problems and how the ARA team solved or is trying to solve them. We will also discuss mathematics and applied problem solving in Healthcare with a focus on career prep and projections.

    Examples: From simple rules to machine learned algorithms trained on features generated from topological data analysis, the ARA data scientists and solution architects saved over 0 million in 2015 from fraud, waste, errors and abuse. Your likelihood to complain is being calculated such that once authenticated, you can be routed to a more appropriate call center agent--improving the consumer experience.

    Jason Haupt is the Associate Director of Healthcare Analytics Technology for UnitedHealthcare's Advanced Research and Analytics team. He received a B.S. in Physics and B.S. in Astrophysics from the University of Minnesota-Twin Cities. He received his PhD in Experimental High Energy Particle Physics awarded by the University of Minnesota-Twin Cities. His thesis work involved 6 years of work with the CERN proton collider. Jason then moved into healthcare as a statistician developing and deploying models for a large provider organization and led teams of analysts as a manager. As a 6th generation Minnesotan, Jason looks forward to the long cold winters each year.
  • Modeling and Analysis of Electric Power Systems Problems: Bridge the Gap Between Research and Industry

    Dongbo Zhao, Eaton Corporation
    February 12, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    Electric power grid has always been crucial to the healthy flow of lives and economy of a nation. Multiple concerns in operating power systems have been raised up and discussed throughout the entire energy industry, including policy-makers and end-point consumers. Power systems span from distant generation site all the way towards the consumption point, which are typically classified as generation, transmission, and distribution grids. Recent calls for smart grid and grid modernization have also elevated strong interest on microgrid and demand management.

    Researchers play significant role in the advancement of power system technologies given their capability to model and simulate existing industry problems. In this seminar, Dongbo is going to cover the top concerns and problems in electric power industry, and to point out the link between academic research and engineering needs.

    Brief Bio: Dongbo Zhao is a Lead Engineer in the Control Systems and Technologies group in Corporate Research and Technology division in Eaton Corporation, located in Eden Prairie, MN. He leads and participates in several R&D projects dealing with distribution automation, grid control and simulation. He obtained his Ph.D in power system reliability analysis area from Georgia Institute of Technology.

  • Building the General-purpose Factory

    Haldean Brown, Plethora
    February 19, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    The last decade has seen the popularization of many prototyping technologies, bringing fabrication methods like 3D printing and laser cutting to the masses with easy-to-use user interfaces, fast turnaround and low prices. Many traditional methods of manufacturing have not benefitted from the same improvements, but are still an essential part of the prototyping process for engineers around the world. Plethora is attempting to bring the same feeling of effortlessness to these tried and true methods; we are doing so by building a factory that is powered by software systems, capable of automatically manufacturing parts using processes that can guarantee high accuracy and quick turnaround.

    In this talk, I will give an overview of the problems that the Computational Geometry team has solved in the process of creating a system that is capable of automatic manufacturing using CNC mills with minimal human involvement. These problems range from P to NP-hard, and from optimal graph traversal to complex surface approximation.

    Haldean Brown is a Senior Computational Geometer at Plethora, and leads Plethora’s Toolpath team, which develops the software that takes a 3D model and outputs instructions for Plethora’s factory. Prior to Plethora, he worked at Google on the Android Wear operating system and on Android at Home, a connected home project. He holds a Bachelor of Science in Computer Science from Columbia University, where he conducted undergraduate research in the Columbia Computer Graphics Group.

  • The Role of Mathematical Modeling and Optimization in Power Grid

    Mihai Anitescu, Argonne National Laboratory
    February 26, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    The electrical power grid (the electricity transmission and distribution system) is one of the greatest and most complex engineering achievements of the 20th century. However, it is also at the center of massive changes in the way we create and consume energy that are brought about by many drivers, including an increasing usage of renewable energy and natural gas. Moreover, it exhibits persistent conceptual difficulties that, while generally successfully contained by engineering practice, have never been fully resolved. In this talk we discuss some of these challenges and the important role that mathematical modeling and optimization can play to solve them. We will argue that in some cases a change of the problem framework may be desirable and that this may be done while keeping the solution computationally achievable. We will outline a number of existing and emerging fundamental research challenges and discuss some recent promising avenues in the area. A distinguishing feature of power grid applications is that optimization is ubiquitous and that it must accommodate simultaneously multiple complexity drivers. These include not only discrete variables, non convexity or stochasticity, but also ordinary and, with the increased usage of natural gas, partial, differential equations. We will discuss the productivity and performance implications of this fact for the modeling and computational environments.

    I have been a Senior Computational Mathematician in the Mathematics and Computer Science Division at Argonne National Laboratory since 2013, and a Computational Mathematician between 2002-2013. I have been a Professor with tenure in the Department of Statistics at the University of Chicago since 2012, jointly appointed with Argonne. Previously, I had been a part-time Professor of Statistics at the University of Chicago since 2009. Between 1999 and 2002 I was an Assistant Professor of Mathematics at the University of Pittsburgh. From 1997 to 1999 I was the Wilkinson Fellow in scientific computing at Argonne. I am currently an associate editor for Mathematical Programming series A and B, SIAM Journal on Optimization, SIAM Journal in Scientific Computing, SIAM-ASA Journal on Uncertainty Quantification and a software editor for Optimization Methods and Software. I specialize in numerical optimization, numerical analysis, and uncertainty quantification. I have advised two Ph.D students at Pitt, two Masters students at the University of Chicago, 22 summer interns at Argonne, and 18 postdoctoral scholars at Argonne. I have sponsored, supervised and mentored 6 full-time scientific employees at Argonne (5 junior scientists and one predoc), all of which hold currently scientific positions or are enrolled in Ph.D programs. I have been a lead investigator in several competitively awarded grants whose total funding amount exceeds 20 million dollars. My work has been primarily in the are of numerical optimization, numerical analysis and uncertainty quantification and their applications in areas related to energy.

  • Group-theoretic Algorithms for Matching Problems with Applications to Computer Vision

    Deepti Pachauri, 3M
    March 4, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    Matching one set of objects to another is a fundamental problem in computer science. In computer vision it arises in the context of finding the correspondences between multiple images of the same scene taken from different viewpoint. In machine learning one often needs to align examples before a meaningful similarity measure can be computed between them. These problems often reduce to some form of combinatorial search problem, various classes of which are polynomial time solvable whereas many others are NP-hard. Therefore, most successful methods take heuristic approach with no optimality guaranteed. The fundamental difficulty in solving matching problems is that the solution space is exponential in size. In her thesis, Deepti considered a different approach for these intractable problems. She leveraged the algebraic structure of the solution of a matching problem i.e., a permutation. A permutation of order n is a candidate in the Sn, called the Symmetric Group of degree n. The high degree of regularity of the symmetric group allows us to unleash a wide range of mathematical tools on matching problems. Her dissertation studied matching problems from the following perspectives: 1) How to take advantage of the fact that any finite group supports a notion of Fourier transformation, and hence Harmonic Analysis; 2) How to extend the notion of regularity in output space of matching problems to solve multi-way matching problem; 3) How to organize matching instances and construct meaningful metric for such data. Proposed approach is fully general and equally applicable to matching problems in domains other than computer vision such as social networks.

    Deepti Pachauri is a Research Scientist in the Advance Technology Group at 3M Corporate Research Lab since April 2015. Her research interest and 3M projects lies at the intersection of mathematics, computer vision, 3D analytics, and machine learning. Prior to 3M, she obtained PhD from the Department of Computer Sciences - University of Wisconsin Madison. In her PhD thesis, she explored the role of geometry, group theory, and statistics in matching problems that are ubiquitous in computer sciences, ranging from computer vision to networks to bioinformatics. Her thesis work was instrumental to a successful NSF award. Pachauri published in highly cited conferences including ICML, NIPS, JMLR and IEEE TMI. She has been on the program committee and reviewer for many international conferences including CVPR, ECCV, ICCV, NIPS, IJCAI, Medical Physics. Indian Institute of Technology - Delhi and the University of Wisconsin Milwaukee are her alma maters where she studied Physics.

  • Lecture

    Lanhui Wang, MaxPoint
    March 11, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract forthcoming.

  • Case Outcome Detection: Challenges and Methods

    Thomas Vacek, Thomson Reuters
    March 25, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract

    The Case Outcome Detection task is the problem of automatically determining the outcome of a lawsuit (dismissal, jury verdict, etc.) based on electronic docket entries created and kept by the respective court. A solution to this problem has important commercial value because it allows service providers to collect valuable statistics of outcome distributions by multiple dimensions (i.e., judge, law firm, attorney, company) We will introduce the data sources for the task and describe how we model the docket progression as a linear-chain conditional random field (CRF). We will show that the CRF approach considerably increases outcome accuracy compared to prior state of the art.
  • Lecture

    Tim Denison, Medtronic
    April 1, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract forthcoming.

  • Lecture

    Adama Tandia, Corning Incorporated
    April 22, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract forthcoming.

  • Lecture

    Tyler Whitehouse, Quantitative Scientific Solutions, LLC
    April 29, 2016 1:25 PM - 2:25 PM
    Lind 305 [Map]

    Abstract forthcoming.

Previous Industrial Seminars

Connect With Us:
Go