Project

Last updated: September 24, 2013

Big Picture

What will you be doing?

For this project, you will:

• read an article in a research journal that uses a stochastic process to model and study a real-world system, and
• write a review of the article that summarizes and critiques the article's approach to studying the system.

What's the point of this project?

By doing this project, I want you to:

• discover an application of stochastic processes that you find interesting,
• gain some experience with reading scientific literature, and
• practice technical writing — a skill you will use well beyond your time at USNA.

Who do you work with? When is the project due?

• You must work in teams of 2 (one team may be just 1).
• A hard copy of your first draft is due in class on Tuesday November 19.
• I will make comments on your drafts and return them to you on Tuesday November 26.
• A hard copy of your final draft is due in class on Thursday December 5.
• If you are finding your own article, you must get my approval by Thursday, October 31. In addition, please include a hard copy of your article when submitting your first and final drafts.

Details

What article should you read?

You may choose to read one of the following articles:

• S. R. Agnihothri, P. F. Taylor. Staffing a centralized appointment scheduling department in Lourdes Hospital. Interfaces 21(5): 1-11, 1991. [article on JSTOR]
• P. Kolesar. Stalking the endangered CAT: a queueing analysis of congestion at automatic teller machines. Interfaces 14(6): 16-26, 1984. [article on JSTOR]
• J. Meredith. A Markovian analysis of a geriatric ward. Management Science 19(6): 604-612. [article on JSTOR]
• D. G. Morrison and R. D. Wheat. Misapplications reviews: pulling the goalie revisited. Interfaces 16(6): 28-34. [article on JSTOR]
• J. M. Steele. Models for managing secrets. Management Science 35(2): 240-248. [article on JSTOR]
• P. Zipkin. Mortgages and Markov chains: a simplified evaluation model. Management Science 39(6): 683-691. [article on JSTOR]

I chose these articles because

• they represent a wide range of applications, and
• you should be able to understand these articles with a reasonable amount of effort.

Some of these articles will require reading ahead in the course.

You are welcome to find your own article for this project. Google Scholar is a pretty good place to start. If you choose to find your own article, you must get my approval first (see above).

What should you write in your review?

Your review should address the following:

• What real-world system are the authors studying? Why is this study important?
• How do the authors model this system as a stochastic process? If they use a stochastic process model that we did not cover in class, provide an algorithmic description (using our general framework in Lesson 7).
• With their model, what assumptions do the authors make about the system they are studying? Are these assumptions reasonable? Why or why not?
• How do the authors validate their model using data from the real-world system, if they do at all?
• How can the authors' model be improved?
• What insights and conclusions do the authors draw from studying their model?

Guidelines and suggestions

• Reading a scientific article takes a lot of time. START EARLY. Don't let the relatively small page counts fool you. The math in research articles is often not cleanly presented like in a textbook. Furthermore, you may need to seek some additional sources to fully understand what is going on — either by reading ahead in our textbook, or looking at other textbooks.
• Focus on making your review well-written and concise. Note: "concise" doesn't mean "no mathematical details." There is no minimum or maximum length for your review. Your review should contain just enough detail so that any of your classmates can read it and get a good idea of what's going on in the article.
• Cite all your sources (at minimum, this should include the article you're reviewing). You can use whatever citation style you like (e.g. MLA, APA); just be consistent.
• Proofread, proofread, proofread.

How will you be graded?

Your grade will be based on the following criteria:

• Submitting a complete first draft (20%)
• Demonstrating a reasonable understanding of the article (20%)
• Addressing the questions above (20%)
• Writing clearly and concisely, and in a well-organized fashion (20%)
• Meaningfully revising and improving your first draft (20%)

Credits

The idea for this project was borrowed from the course project for Math 4740 at Cornell.