Industrial Projects

2009-2010 Projects

Project Pele: Humanoid Robotic Programming A Study in Artificial Intelligence

Sponsor: Huazhong University of Science and Technology
Advisors: Jon Abraham, Yiming Rong (ME)
Students: Neil Nanisetty, Wade Mitchell-Evans, Erik Fajardo, Joseph Schlesinger

In the ever changing world of technology, the humanoid robot has been a constant member of science fiction culture. Our project goal was to develop a humanoid robot capable of independently displaying effective soccer skills. We divided the tasks into two teams; one designed a ball kicking robot program while the other designed a path tracking robot program. After each group completed their four major objectives, we had created a superior program than its predecessors. Using our optimized code as a foundation, another group can further develop these robot programs to demonstrate even more humanlike soccer skills.

Information Sharing at Bank of America

Sponsor: Bank of America
Advisors: Jon Abraham, Daniel Dougherty (CS), Arthur Gerstenfeld (MG)
Students: Marco Angulo, Burcu Bora, Kimberly Gallagher, Chao Zhang

This project, conducted concurrently in New York and London, sought to improve the information sharing practice for the trading support teams in the Global Credit Products division of Bank of America, which has gained vital importance due to the increased volume of business following the merger with Merrill Lynch. After an extensive assessment of the previous practice, detailed recommendations were made accordingly to address a formal knowledge management approach which would enhance the effectiveness and efficiency of information sharing. Finally, a new platform was implemented with a clearly defined structure and various features in order to streamline the information sharing procedure.

Analysis of Fund Mapping Techniques for Variable Annuities

Sponsor: Towers Perrin
Advisor: Jon Abraham
Students: Grant Fredricks, Erin Ingalls, Angela McAlister

Our project tested the accuracy of projected mutual fund returns (via fund mappings) relative to the actual performance of the mutual fund. To achieve this, we performed Monte-Carlo simulations using geometric Brownian motion. Utilizing Excel's VBA, we generated many scenarios to draw conclusions about fund mapping reliability. We modified parameters within the simulation in order to isolate factors that affected the fund mapping's accuracy, and analyzed the specific effects of these variables to provide our sponsor, Towers Perrin, with recommendations.

Predicting Policyholder Behavior and Benefit Utilization: An Analysis on Long-Term Care Insurance

Sponsor: Ability Resources, Inc.
Advisors: Jon Abraham, Helen Vassallo (MG)
Students: Jie Bai, Ashleigh Smeal, Heather Standring, Xinyi Zhang

In order to better serve their customers, a project to create a methodology for identifying variables that could indicate future long-term care insurance usage was commissioned by Ability Resources, Inc. As a basis for constructing a predictive model, tools such as SAS and Excel were implemented. A k-means clustering algorithm in SAS was utilized to group policyholders with similar characteristics, and a performance evaluation was executed in Excel. Together, these processes created a tool that determined the impact each characteristic had on policyholder benefit utilization. The validity of the process was assessed by applying it to supplemental data generated by the team. After several trials, the Variable Identification Procedure proved accurate.

Worker's Compensation

Sponsor: The Hanover Insurance Group
Advisor: Jon Abraham
Students: Joshua Davis, Steven Ditullio, Nicholas Vine, Thomas Whiting

This project was designed to attempt to predict the insurance cycle, specifically for Workers' Compensation. A process was created which involved regressing premium values against external economic indicators in order to predict future premium values. We tested the data several times using different modifications and groupings of the provided data. Our results were then compared against the unused data points. Finally, we recommended certain possibilities for further analysis or testing.

Maintained by webmaster@wpi.edu
Last modified: Aug 04, 2016, 15:31 UTC