Search Content

Matching Items (3)

Filtering by

All Subjects: Statistics

Prioritizing Projects on Time and Cost Savings for a more Efficient Manufacturing Process of a Semiconductor Company

Description

Over the course of six months, we have worked in partnership with Arizona State University and a leading producer of semiconductor chips in the United States market (referred to as the "Company"), lending our skills in finance, statistics, model building, and external insight. We attempt to design models that help predict how much time it takes to implement a cost-saving project. These projects had previously been considered only on the merit of cost savings, but with an added dimension of time, we hope to forecast time according to a number of variables. With such a forecast, we can then apply it to an expense project prioritization model which relates time and cost savings together, compares many different projects simultaneously, and returns a series of present value calculations over different ranges of time. The goal is twofold: assist with an accurate prediction of a project's time to implementation, and provide a basis to compare different projects based on their present values, ultimately helping to reduce the Company's manufacturing costs and improve gross margins. We believe this approach, and the research found toward this goal, is most valuable for the Company. Two coaches from the Company have provided assistance and clarified our questions when necessary throughout our research. In this paper, we begin by defining the problem, setting an objective, and establishing a checklist to monitor our progress. Next, our attention shifts to the data: making observations, trimming the dataset, framing and scoping the variables to be used for the analysis portion of the paper. Before creating a hypothesis, we perform a preliminary statistical analysis of certain individual variables to enrich our variable selection process. After the hypothesis, we run multiple linear regressions with project duration as the dependent variable. After regression analysis and a test for robustness, we shift our focus to an intuitive model based on rules of thumb. We relate these models to an expense project prioritization tool developed using Microsoft Excel software. Our deliverables to the Company come in the form of (1) a rules of thumb intuitive model and (2) an expense project prioritization tool.

ContributorsAl-Assi, Hashim (Co-author) / Chiang, Robert (Co-author) / Liu, Andrew (Co-author) / Ludwick, David (Co-author) / Simonson, Mark (Thesis director) / Hertzel, Michael (Committee member) / Barrett, The Honors College (Contributor) / Department of Information Systems (Contributor) / Department of Finance (Contributor) / Department of Economics (Contributor) / Department of Supply Chain Management (Contributor) / School of Accountancy (Contributor) / School of Mathematical and Statistical Sciences (Contributor) / Mechanical and Aerospace Engineering Program (Contributor) / WPC Graduate Programs (Contributor)

Created2015-05

A Fortune 100 Technology Company Collaborative Thesis: Third-Party Services \u2014 Optimizing Headcount

Description

The object of the present study is to examine methods in which the company can optimize their costs on third-party suppliers whom oversee other third-party trade labor. The third parties in scope of this study are suspected to overstaff their workforce, thus overcharging the company. We will introduce a complex spreadsheet model that will propose a proper project staffing level based on key qualitative variables and statistics. Using the model outputs, the Thesis team proposes a headcount solution for the company and problem areas to focus on, going forward. All sources of information come from company proprietary and confidential documents.

ContributorsLoo, Andrew (Co-author) / Brennan, Michael (Co-author) / Sheiner, Alexander (Co-author) / Hertzel, Michael (Thesis director) / Simonson, Mark (Committee member) / Barrett, The Honors College (Contributor) / Department of Information Systems (Contributor) / Department of Finance (Contributor) / Department of Supply Chain Management (Contributor) / WPC Graduate Programs (Contributor) / School of Accountancy (Contributor)

Created2014-05

Contract Year's Effects on NBA Salary Caps and What Statistics Can Identify These Players

Description

The National Basketball Association is one of the most popular and most profitable sports leagues in the entire world, and with stars like Giannis Antetokounmpo, Nikola Jokić, and Luka Dončić, it continues to expand its international reach. In the past decade this has meant that the salary cap has continued to increase considerably. From 2013 to 2017 the salary cap increased by about $40.5 million from around $58.5 million to $99 million meaning there was an extra $1.2 billion worth of cap space to fill. All this new cap space created a perfect storm for numerous players to be overpaid. Many saw the performance of these overpaid players as a part of the contract year phenomenon where a player performs better before a new contract and then after receiving their new contract, their performance deteriorates. The purpose of this research is twofold. First, it looks at whether the contract year phenomenon has been present in the NBA since 2015. After that it looks to find what statistics are the best predictors for performance based on their positions. This was done through various statistical analysis techniques such as T-tests and piecewise regression. Box score statistics like point, rebounds, and assists as well as advanced metrics like player efficiency rating, usage percentage, and true shooting percentage were utilized in this study. The results indicated that the concept of the contract year phenomenon was present in the players sampled. However, rather than contract year only being for players who increased their performance in the previous year, it is a more general phenomenon. Also, there was major differences in the statistics that predicted performance. The biggest of these was the importance of usage percentage rather than points and that centers had the least predictors, most likely due to the evolution in the play of centers.

ContributorsUmland, Matthew (Author) / Eaton, John (Thesis director) / McIntosh, Daniel (Committee member) / Barrett, The Honors College (Contributor) / School of Accountancy (Contributor) / WPC Graduate Programs (Contributor)

Created2022-05