If the project isn't interpretable as a simulation or LP/IP problem,
you may get reduced or no credit. Data must be real
data, not just something you made up or a textbook example. Explain
where you got the data or you may not get credit for this assignment.
There is no special length requirement. Typically 1-4 pp. brief description
of the question/problem, source of data, approach, results, significance.
Real data, Solver optimization or monte-carlo simulation such
as EXTEND or EXCEL random numbers (usually with macros).
don't just redo one of the exercises. No more fast food diet problems.
Should be something reasonably novel (something
other than nutritional value of fast food or other rework of the exercises).
Due the last day of class. None are accepted late.
Use google to find work others have done in solver as a starting place.
Try the web or go collect data someplace like at work, a bank, a grocery
store... e.g. try playing with EXTEND like a video game to investigate
effects of breaking up processes to multiple steps. Some stores have implemented
is a link to a google search showing a few of these. One very interesting
application of simulation
is spread of infectious diseases.
Food stamp debit cards are issued at the beginning of the month in Michigan
, so some grocery stores have 50% of their business in the first 9 days
of the month npr.org search food stamps. Search
google.com "food stamps percent population" for more data, and see the
Bureau of Labor statistics data. What would it be worth to stagger food
stamp disbursements to twice a month, versus inventory holding costs in
the grocery stores? This is a big issue as between 5-10% of the population
is on food stamps, as high as 50% in some locales.
Use similar searches to find applications for linear programming, such
as Solver or POM WIN. Part of the exercise is to identify novel applications.
that's why I won't grade it very rigorously as long as you have real data
and use one of these tools.
If you really have a hot project in mind, consider expanding it
a little to make it presentable at the Undergraduate
Research Symposium on April 27 2007.