Last updated:
- March 17, 2014 — Original version.
Overview
Your goal: write a report that
- describes the methodology you developed and used to solve your
capstone project problem, and
- describes and interprets the results you have obtained.
This will eventually form the bulk of the final capstone report you
will submit at the end of the semester.
What should this report contain?
The best way to present your methodology and results will depend on the
exact nature of your project. Your report should address the following in some
shape or form. See below for examples.
- Describe your problem mathematically. In particular:
- Define all the input parameters to your problem as symbols.
- Explain what constitutes a feasible solution to your problem: i.e. describe the constraints.
- Describe the objective of your problem.
- Describe your methodology for solving your problem. Some or all
of the following may apply:
- Describe how you collected and computed data for the problem
(i.e. how did you come up with the numbers associated with all
the input parameters you defined above?).
- Give a mathematical programming formulation (e.g. linear
program, integer linear program) of your problem. Be sure to
describe the meaning of each component of your formulation in
words.
- Give pseudocode for the algorithms you wrote to solve your
problem. If you have a lot of subroutines, a flowchart may be
helpful.
- Describe the computing environment and the software that you
used.
- Discuss and interpret the results you obtained from carrying out your methodology.
For this report, discussing and
interpreting preliminary results is OK.
- According to your methodology, what is the solution to your
problem? Don't just report raw numbers — what does the solution
mean in the context of your problem?
- Carry out sensitivity analysis: how does your solution
change when the input parameters change? Is your proposed
solution robust or sensitive to these changes? What
implications does this have in the context of your problem?
References and citation style
Use the same citation style as the one you used for your problem
statement and literature review:
- References should be listed in alphabetical order at the end of
the report.
- List journal articles like this:
Y. Linde, A. Buzo and R.M. Gray.
An algorithm for vector quantizer design.
IEEE Transactions on Communications,
28(1):84-95,
1980.
- List books, pamphlets, and research reports like this:
H. Abut, editor.
Vector Quantization.
IEEE Press,
1990.
- List papers in a bound collection, such as conference proceedings like this:
M. Fanty, P. Schmid, and R. Cole.
City name recognition over the telephone.
In Proceedings of the International Conference on Acoustics, Speech and Signal Processing,
volume 1,
pages 549-552,
Minneapolis, U.S.A.,
1993.
- Refer to references by the authors' last names and year of
publication in round brackets, like (Linde et al. 1980) or (Abut
1990). If you want to refer to a specific page or result, you can do
so like (Abut 1990, p. 51) or (Fanty et al. 1993, Figure 2.3).
Suggestions
- Examples: take a look at
certain parts of this article
to get an idea of what the language in your report should look like:
in particular, look at Section 2 for an example of a mathematical
problem description, Sections 3.1 and 3.2 for examples of presenting
and explaining mathematical programming formulations, and Section 3.3
for examples of algorithms in pseudocode and presenting results.
- As usual, focus on making your report organized and well-written.
- Also: proofread, proofread, proofread.
How will you be evaluated?
Your report will be evaluated based on the following criteria:
- Submitting a complete draft
- Giving a complete and correct mathematical description of your problem
- Describing your methodology thoroughly
- Presenting the (preliminary) results you obtained from your methodology,
and discussing and interpreting these results in a thoughtful manner
- Writing clearly, concisely, and eloquently, and in a well-organized fashion