GRADUATE LEVEL PROJECTS

 

 

Masters: Management Information Systems

Masters: Supply Chain Management

1. Individual Personal Awareness and Group Project Paper

2. SAP Supply Chain Management

2. Association Rules and Clustering.

3. Using Data Mining for Direct Marketing

4. Data warehousing strategy and planning for public/government and private sectors

5. Software Project Management covering knowledge areas from the PMIBOK

 

 

 

 

1. Individual Personal Awareness and Group Project Paper

 

Semester: Spring 2008

 

Team Size: 5

 

Team Members: Akshay Ajmani, Mohit Malhan, Jeff Batson , Ricky Lee, Cheng Hung , manyun hsa , Li-Ying Cheng.

 

Guided by: Prof. David Ford.

 

The group project report involves developing a paper from the perspective of a self-analytic group that focuses on its developmental processes and growth during the semester. The developmental processes are to be the main focus of the paper because more intimate knowledge of each other will be required to complete the project paper from this perspective. The group needs to generate its own data around which it will build its paper. These data will result from a sociometric analysis which the group is to complete with respect to two roles — that of task leader and lunch partner. The project report should focus on answering the following questions: To what level of development has the group progressed? What is the prognosis in the near term for the group if it were to continue its life beyond the end of the semester and this class? What factors have facilitated or hindered the participation of group members in helping the group to coalesce and/or gel? How do you explain the group’s perception of its overall performance as well as performance in specific areas based on the mean scores assigned by each member? In what ways does the Kendall Life Languages Profile help explain your group’s developmental process? All of this discussion should essentially explain why the sociometric diagrams look the way they do. This analysis is to be accomplished by using the templates found on the course WebCT website. These templates include: (1) conceptual framework, (2) sociometric rating instrument, (3) relationship chart, (4) relationship ratings, totals, reciprocal relationships, and group cohesion index, (5) diagnosing team effectiveness rating form, and (6) ratings of satisfaction and task effectiveness. Templates #3 & #4 are to be developed for each role of task leader and lunch partner.

 

2. SAP Supply Chain Management

 

Semester: Summer 2008

Team Size: 1

Guided by: Prof. Gene Deluke.

 

 

3.  Association Rules and Clustering (Business Intelligence)

 

Semester: Fall 2008

 

Team Size: 4 *()

 

Team Members: Mohit Malhan, Akshay Ajmani, Nimish Patil, Akhil Jain

 

Guided by: Prof. Eric Zheng.

 

The goals of this exercise are: 1) familiarize yourself with your first-ever data mining package – XLMiner; 2) use it to mine association rules and clusters and 3) analyze the results. Note that a big portion of this project is on identifying business opportunities and analyzing the results, the type of tasks meant for business graduates (think otherwise computer science students). The data (available at Webct for downloading) we use is from a major web-intelligence vendor for the online travel industry. Each instance records a customer’s web purchasing history: what web sites she has purchased from. In addition, the data also consists of demographic information and the dollar amount the customer spent. Your job is to use association rules and clustering implemented in XLminer to solve the business problem you identified.

 

This is an open project. You are supposed to come up with your own business problem and apply XLminer to solve it. Specifically, you need to understand the dataset and figure out what the data could tell you. Suppose you are working for a firm appearing in our dataset. You are asked to identify business problems (opportunities) that association rule and clustering could help solve. In grading, I will pay attention to the significance and interestingness of the problem you identified, as well as how you use the knowledge you learned so far (association rule and clustering) to discover business intelligence for tyhe firm.

 

In the end, you need to submit a report stating clearly what the problem you are solving, the procedures you apply association rule and clustering to solve it, and you recommendations based on the results (of analyzing the data). Through the project, you are supposed to demonstrate you ability of using association rule and clustering in XLminer. The follows are a couple of items I’d like you to include in your report to demonstrate your proficiency.

 

4. Using Data Mining for Direct Marketing

 

Semester: Fall 2008

Team Size: 4

 

Team Members: Mohit Malhan, Akshay Ajmani, Nimish Patil, Akhil Jain

 

Guided by: Prof. Eric Zheng.

 

 

Background: A direct marketing firm mails catalogs to its customer base which consists of 5 million households (in our dataset, we have 2,000 for training and another 2,000 for testing). Catalogs are mailed to all the customers who then either do not respond or respond by ordering items from the catalog. Order amounts vary from a few dollars to hundreds of dollars. The firm has no prior knowledge of which customers would respond to its campaign and how much a customer would spend on an order. Yet, the firm enjoys from a relatively high response rate compared to that of other companies in the industry. The average response to the firm’s campaigns is around ~30% whereas the industry’s advertised average response rate is 22%.

 

In spite of its relative success the firm incurs considerable printing and mailing costs. It distinguishes itself by mailing expensive catalogs. Although the appealing look of the catalog seems to result in higher consumer interest in the firm’s offerings it also incurs higher costs per mailing. The firm calculated that each mailing costs $4 — twice the cost per mailing incurred by competing companies.

 

Much of the firm’s effort in recent years has been to improve its product assortment as well as the presentation of products in the catalogs. However, the recently hired Chief Marketing Officer has proposed to further improve the firm’s performance by identifying profitable customers and distinguishing them from others. Specifically, she suggests targeting consumers that are likely to be responsive and that would order items that would justify the printing and mailing costs. The CEO supports the Marketing Officer’s proposal and decided to dedicate resources to estimate the potential lift in profits that the firm can expect from this initiative. The company is particularly concerned with lapsing customers, i.e., customers who made their last order between 13-24 months ago.

 

An early study was performed by the marketing department to reveal customers’ decision-making patterns. The study shows that consumers make their buying decision in two phases. They first decide whether or not to respond to the catalog, and once they decide to buy from a catalog, they make a separate decision as to what they should buy.

 

Your team was hired to consult with the CEO and the Marketing Officer. The Marketing Officer has provided you with a data set consisting of information about the last campaign. The data describe about customers’ previous responses (or lack thereof) as well as the profit (prior to subtracting the mailing costs) that were accrued from customers’ orders when such were made. Everyone included in this data set had made at least one prior purchase for the catalog.

 

5. PRESENTATIONS & TERMPAPERS

 

Semester: Fall 2008

 

Team Size: 4

 

Team Members: Mohit Malhan, Akshay Ajmani, Nimish Patil, Akhil Jain

 

Name of Guide: Prof. Eric Zheng.

 

Business intelligence was first referred in Tzu s Art of War. According to Tzu in order to win a war one should have complete knowledge of the strengths and weaknesses. This is the core idea behind business intelligence. To compete in the market a company should know its customers and competitors better than anyone else.

 

Most organizations don’t get the information they want from their existing systems. Spreadsheets allow users do exactly what they want but at a very high price. In spreadsheets errors can slip because of calculation mistakes. There is no audit trail on changes and mistakes may not be detected. To make matters worse, spreadsheets are typically not shared across an organization and they are not updated as things change. So decisions are made with old data.

 

BI tools are supposed to take the mechanics out of the box and the analysts don’t have to wait until the IT department gets around to generating the required report. With BI, analysts can slice and dice the data any way they want without asking for help. Since all the BI information is stored on a warehouse rather than one user’s desktop, BI tools should also provide one version of the truth.

 

Some products are already integrated with the high-end ERP systems such as SAP or PeopleSoft – although the latter have their own BI tools. The surveyed products compete with SAP and PeopleSoft because large clients can have multiple ERP systems including SAP, PeopleSoft and others. It seems BI is not that well integrated with mid-market ERP and accounting systems. This will probably change as the BI vendors go after the middle market.

 

6. Data warehousing strategy and planning for public/government and private sectors

 

Semester: Fall 2008

 

Team Size: 1

 

Team Members: Mohit Malhan, Akshay Ajmani, Nimish Patil, Akhil Jain

 

Guided by: Prof. Lou Thomson

 

The first computer, businesses have implemented the latest information technology to improve efficiency and reduce costs. By automating with the latest technology, firms are able to access information more efficiently than heretofore possible. The first companies to implement and achieve benefits through the use of current technology establish a competitive advantage. Soon, however, competitors follow the example and the use of such technology becomes a requirement for entry and survival. Thus, information technology development breeds opportunity for business needs and business needs breed creation of new information technology in a seemingly never-ending cycle. Data warehousing is part of this cycle. This article discusses about basics steps involved in the implementation process which will provide us with an aerial view of EDC design for the enterprise based data ware house. Potentials risks involved during the implementation and migrations of data set from one data frame to other will also be discussed along with. Data warehousing strategy and planning for public/government and private sectors will be narrated along with the various implementation successes and failures as well as success stories that were resulted by implementation of business intelligence

 

7. Software Project Management

 

Semester: Fall 2008

 

Team Size: 3

 

Team Members: Mohit Malhan, Akshay Ajmani, Nimish Patil, Akhil Jain

 

Name of Guide: Prof. Mark Thouine

 

5 process groups from the PMIBOK

– Initiation

– Planning

– Executing

– Monitoring and Control

– Closing

 

9 knowledge areas from the PMIBOK

– Time Management

– Cost Management

– Scope Management

– Quality Management

– Procurement Management

– Integration Management

– Risk Management

– Human Resource Management

– Communication Management

 

Systems development lifecycle

– Requirements Management

– Design Management

– Construction (Development) Management

– Testing Management

– Maintenance Management

– Configuration Management

– Process Management

– Other