You are here

PIC Math Student Recognition Conference Oral Presentation Abstracts

Student Oral Presentations: Saturday, 8:30 a.m. – 10:30 a.m., International Ballroom South

The PIC Math (Preparation for Industrial Careers in Mathematical Sciences) program aims to prepare mathematical sciences students for industrial careers by engaging them in research problems that come directly from business, industry, or government. During the spring 2017 semester, mathematical sciences undergraduate students at 67 U.S. universities and colleges were enrolled in a PIC Math industrial mathematics and statistics research course. Each student team worked on a research problem and submitted a written report and video solution to the problem to the PIC Math student research competition. Several student teams will give presentations of their problems and solutions during this session. PIC Math is a program of the MAA and SIAM supported by NSF funding (DMS-1345499). See http://www.maa.org/picmath

Organizers:
Michael Dorff, Brigham Young University
Suzanne Weekes, Worcester Polytechnic Institute

Opportunistic Rebalancing

8:30 a.m. - 8:40 a.m.
Alana Danieu, Emily Kaegi, Annie Shapiro, Daniel Weithers, Carleton College

Determine an optimal stock portfolio rebalancing strategy based on historical market performance

A Mathematical Model based on IC50 Curves to Predict Tumor Responses to Drugs

8:45 a..m. - 8:55 a.m.
Catherine Berrouet, Jake Nadulek, Sunil Giri, Emmanuel Fleurantin, Florida Atlantic University

The standard measure of the drug dose needed to kill (or inhibit the growth of) half of the tumor cell population is called the IC\(_{50}\) concentration. To determine the IC\(_{50}\) value, the cells are grown for 72 hours (or 96 hours) in separate dishes, each with the increasing concentrations of the drug. However, in the Petri dishes all cells are equally well exposed to the drug. The goal of this project is to develop a mathematical model to investigate how to use (or improve) the IC\(_{50}\) approach to control the growth of 3D tumors.

Classifying Queries Based on the North American Product Classification System (NAPCS)

9:00 a.m. - 9:10 a.m.
Ashley Sexton, Tianna Burke, Howard University

Every five year the US Census Bureau conducts a census of the goods and services available in the country. An important challenge in this task is to provide the producer and consumer with quick and accurate access to information they might require. The aim of this project is to classify words and phrases that fall into specific categories based on the North American Product Classification System. Using vector representation we devised several ranking algorithms to find the best categories for a given word/phrase.

Ensuring the Insurer: A Modern Statistical Approach to Estimating Unpaid Losses

9:15 a.m. - 9:25 a.m.
Laura Farro, Courtney Taylor, Sam Kunkler, Northern Kentucky University

This project details a modern statistical approach to estimating loss reserves. Working with data sets supplied by American Modern Insurance Group, new models are developed and tested against standard actuarial techniques.

Assembly Line Efficiency at a Large Manufacturing Plant

9:30 a.m. - 9:40 a.m.
Levi Nicklas, Jacob Kautz, Shippensburg University

Develop a mathematical model, and associated computer-based tool, to identify efficient assembly line structure at a large construction manufacturing facility.

Predicting Customer Insurance Type

9:45 a.m. - 9:55 a.m.
Megan Sharp, Benita Beale, Matthew Lyons, University of Washington, Tacoma

A health insurance company has a large dataset from a survey including health insurance plan, demographics, media habits, and many other variables. They would like to understand common characteristics of healthcare consumers---is there a way to segment these consumers based on criteria that could help the healthcare company make marketing and strategy decisions?

Profit Optimization

10:00 a.m. - 10:10 a.m.
Quinn Burzynski, Lydia Frank, Zac Nordstrom, Jack Wolfe, University of Wisconsin, La Crosse

The students analyzed past quoting data for a local industrial parts supplier. They utilized machine learning techniques to identify most significant factors for quoting success. They discovered optimal discounts per product categories to maximize profit when selling to new customers.

Analyzing New Health Care Placement of Mercy Health Facilities

10:15 a.m. - 10:25 a.m.
Natalie Halavick, Leah McConnell, Khang Nguyen, Sara O’Kane, Youngstown State University

Mercy Health asked for recommendations regarding the placement of a new inpatient or outpatient health care facility in the Youngstown area. Given data on health care utilization and projections for health care usage for the next five years, the students developed a model and provided potential locations and lines of service for such a facility.

Year: 
2017

Dummy View - NOT TO BE DELETED