Evaluating Accounts Payable Policies and Procedures
Student Researchers: Duc Chu, Shiladitya Roy Chaudhuri, Maaz Khan
Faculty Advisor: Lisa Verdon (Economics)
The team worked with the College’s Accounts Payable (AP) department. The project was created to increase the efficiency of day to day to operational activities of the AP department through the use of process map. By analyzing the process map the team was able to provide recommendations and suggestions that would help the AP department to reduce process cycle time, decrease defects, reduce costs, reduce non-value-added steps, and increase productivity.
Monitoring and Evaluation Framework for Adivasi Development Network
Student Researcher: Sadaf Asrar
Faculty Advisor: Jim Burnell (Economics)
The purpose of this project was to create a Monitoring and Evaluation (MandE) Framework for Adivasi Development Network (A.D.N.), a Non-Profit Organization which works with the indigenous people in Eastern India on basic development issues through collaboration with Adivasi non-profits and leaders in Eastern India for the advancement of the Adivasi, the indigenous people of India. The MandE Framework is created using the Logical Framework Approach employing Logframe Matrices and MandE Matrices to provide a step-by-step guideline on how to conduct MandE of multidimensional development projects.
Analyzing Mobile Banking in Microfinance and Developing an SMS Based Program to Facilitate Microfinance Institutions using Mobile Banking with their Clients
Student Researchers: Chris Miller and Micah Caunter
Faculty Advisor: Jim Burnell (Economics)
The purpose of our research was 1) to examine and analyze the market opportunity associated with implementing Mobile Banking in Microfinance 2) to develop and implement an SMS-based mobile banking program that MFIs can use to conduct mobile-based interactions with their clients. Specifically, our team examined the current practices of Microfinance and Mobile Banking by financial institutions across Africa and Latin America.
The market research also entailed researching the potential of developing mobile software that would be able to better serve the “unbankable” population in developing countries. Accordingly, we developed an aspect of a larger mobile banking specification that aims at using mobile technology to consolidate financial services. The software will allow MFIs to manage the messages that are transmitted between their clients.
Modeling the tan Delta Spectrum
Student Researchers: Norman Israel, Hannah Dauber, Robert Taylor
Faculty Advisor: John David (Mathematics)
In rubber, the Tan Delta spectrum is the ratio between loss and storage modulus which contains information about the mechanical properties of rubber. In 1986, K.H Nordsiek found that various parts of the tan Delta spectrum were diagnostic of various tire performance features related to the tread compound. He believed that ideal tread compound performance could be realized through mixtures of rubbers. The AMRE research team developed a model to predict the tan Delta spectrum from the experimental data on the tire ingredients. The team used Neural Networks to do this, repeatedly creating and training various networks to get better prediction each time, finally creating an optimal predictive model. The team was also able to use the networks to identify the important tire ingredients for tan Delta value. This research is important to Goodyear because it improves on previous approaches they used for modeling the spectrum. This can reduce the amount of time they spend on testing. The work done by this research team also provides valuable information to guide future research done by Goodyear (such as with the ranking of the ingredients according to importance). The work on confidence bands for the prediction also offers Goodyear new predictive ability.
Surface Competition of Selected Cure Chemicals, Antioxidants, Silanes and Water onto Model Carbon Black, Zinc Oxide and Amorphous Silica Surfaces
Student Researchers: Melissa Venecek, Pam Wales
Faculty Advisor: Sarah Schmidtke (Chemistry)
This project was developed as a computational chemistry project to compute the relative sorption energies and model the structures of selected rubber chemicals on filler surfaces. The compounds are used in the vulcanization process for producing tires. The main goal was to use free energy of sorption as a measure of the surface competition by each rubber chemical. The chemicals have an effect on the rubber crosslink density, filler surface and aging characteristics of the rubbers. Each of these rubber compounds sorbs onto a filler, but unfortunately there is not much knowledge as to how the rubber chemicals interact with the fillers. Throughout this research there were four specific milestones: obtaining the lowest energy geometry of the chemicals sorbed onto filler surfaces, computation of relative solvation energies of rubber chemicals, finding a successful computational method for obtaining sorption energies and obtaining the relative energies of absorption for each rubber chemical on each filler surface. All of the calculations in this study were performed using the Gaussian 03 computational chemistry program.
Analysis of Cross-section Geometry of Complex Steel Cords
Student Researchers: Kemar Reid, Andrew Licking, Yanlong Hu
Faculty Advisors: John Ramsay and R. Drew Pasteur (Mathematics)
The purpose of our project was to model geometrically the steel cords used by the Goodyear Tire and Rubber Company. Our clients at Goodyear wanted us to deliver a model (using MATLAB) that could display 2-D cross-sections of any specified steel cord design, anywhere along its length, in order to analyze the construction. These steel cords for tire reinforcement are created by taking series of filaments and wrapping them together in a helix. After construction, steel cords are placed upon a layer of rubber, while another is laid on top of it in the tire assembly process. The rubber then molds into the gaps in the steel cord, allowing for greater strength in the tire structure. However, when there is overlap in the structure, namely when two filaments try to occupy the same space, they will push out or rotate or in some other way cause an unaccounted for structural deformity, which will then lead to weakness in the tire designs as well. Our model will be used by Goodyear to view, analyze and interpret steel cord constructions used in tire reinforcement.
Wooster Mathematics Knot Theory Research
Student Researchers: Louisa Catalano and David Freund
Faculty Advisor: John Ramsay and Jennifer Roche Bowen (Mathematics)
This project, funded by the GLCA, HHMI, and Sophomore Research Program, developed the foundation for a research program into the field of knot theory. In laying the groundwork for future research in this field, specific areas of knot theory were investigated as potential directions for research; these areas include the arc presentation of knot, tricolorability, virtual knots, and satellite knots. The arc index of a knot was found to be promising, and most of the current research in this area was reviewed. Also, the idea of a Klein bottle knot was invented and preliminary drawings were created.
Tomato Analyzer
Student Researchers: Itai Njanji, Atticus Jack, Joshua Thomas
Faculty Advisor: Jaymie Strecker (Computer Science)
The purpose of our project was to improve Tomato Analyzer, an application that measures shape and color attributes of fruits. Tomato Analyzer is part of the Tomato Fruit Morphology Project, which seeks to understand how genes and molecular networks control fruit morphology. We added several features to Tomato Analyzer. These features include an area that allows the user to add “notes” to the images being analyzed and updates to the Color Analysis feature.
Color Analysis was streamlined and given an expanded interface to make the feature more user-friendly. To accomplish this the user now must only deal with an in program popup window to calibrate the color; this process is much simpler then the previous method that involved exporting values to Excel and running a regression algorithm on them. The program was made more stable and was made to run faster due to the bug fixes made by the team as well. The team also worked on enhancing the boundary detection of the program to allow for more accurate measurements.
Wooster Mathematics Polygon Dissection
Student Researchers: Evan Radkoff and Logan Garrity
Faculty Advisor: Pam Pierce (Mathematics) and Denise Byrnes (Computer Science)
Our project, Circle Squaring 2010, is a continuation of work done by AMRE and HHMI researchers in the past. Our work involved polygon dissections. This was a result of trying to approximate a 1990 paper by Laczkovich on the subject of circle to square decompositions. The past summers saw major results arise including an algorithm to dissect a polygon of any given even number of sides and form a square with the pieces. This summer, the task fell to us to write up the results and create an animation visually explaining our algorithm. The goal was to submit a website with our research findings to the MAA’s online journal, LOCI.
NFL Prediction Using Neural Networks
Student Researchers: Michael Janning and Saif Ahmad
Faculty Advisor: R. Drew Pasteur and John David (Mathematics)
Our research analyzes the ability of a neural network model to predict the outcome of regular season NFL games. This model uses only readily available statistics, such as passing yards, rushing yards, and fumbles lost. A key component of this model is the use of differentials where, for example, the passing yards of one team are compared to the defensive passing yards of the other team. By using principal component analysis and derivative based analysis, we determined which statistics influence our model the most.
We assessed the performance of the model by comparing its predictions to those of media members and the Las Vegas odds makers. We also consider the absolute error in predicting the margin of each game. In both total wins correctly predicted and point spread error, our model performs similarly to the Las Vegas line. Using the second half of the season for our predictions, we obtained an average accuracy prediction of 65.8% for 2006, 72.2% for 2007, 75.8% for 2008, and 68.2% for 2009 over 10 different realizations of the model. The standard deviation for each year was less than 1%.