These are actual student papers that were not designed to be web pages. They may contain historical, grammatical, mathematical, or formatting errors. These papers were graded using the criterion mentioned in the paper directions, and the writing checklist. The test review sheets and the WebCT tests are good indicators of the mathematics that was discussed in class during and/or after each presentation.

There is not a lot of information available about the life and childhood of Margaret H. Wright. One day, however, I am sure there will be plenty written about her life. She is a remarkable woman in the field of mathmatics, science, and technology.

Margaret Wright attended Stanford University where she received her B.S. in Mathematics and her M.S. and Ph.D in Computer sciences. The areas she showed the most interest in were optimization, linear algebra, numerical analysis, scientific computing, scientific applications, and engineering applications.

From 1976-1988 she was a researcher in the Systems Optimization Lab, Department of Optimization Research at Stanford University. From 1988 to the present she is employed at the Computing Sciences Research Center at Bell Labs, Lucent Technologies.

She has been the recipient of several awards in the past decade some of which include:

1993-Distinguished Member of Technical Staff

1994-1998-served on Advisory Committee for the Directorate of Mathematical and Physical Sciences at the National Science Foundation (chaired: 1997-1998)

1995&1996-President of Society for Industrial and Applied Mathematics (SIAM)

1997-elected to National Academy of Engineering

1997-Head of Scientific Computing Research Department

1999-Bell Labs Fellow

She has also recently served on several committees such as the National Research Council, the National Science Foundation, the Department of Energy, and the Scientific Advisory Committee of the Mathematical Sciences Research Institute (MSRI).

She is also an editor for five journals. She is Editor-in-Chief for the SIAM Review and Associate Editor for Mathematical Programming, SIAM Journal on Scientific Computing, SIAM Journal on Optimization, and IEEE Computing in Science and Engineering. She also co authored two books with Phillip Gill And Walter Murray Practical Optimization and Numerical Linear Algebra and Optimization.

"Her work includes two parts: basic research into theory and algorithms in optimization and linear algebra; and solution of real-world optimization problems (Distinguished Member…AT&T Bell Labs)." Her research and solving of practical problems often inspires new problems and research that is often an unexpected challenge but not unaccepted. From the things I have read, Margaret Wright is a very determined woman and never backs down from a challenge because she truly enjoys her job. It seems that she finds every job rewarding because she is able to help in advances toward practical problems by solving the underlying mathematical and numerical issues. She enjoys being able to achieve things that make a difference. She has nothing but good things to say about her job as quoted from an article, "It sounds corny, but magical results happen here," she says. "This place is special; people are absolutely excellent in individual areas of science, and we're encouraged to work together. It's a rare environment (Today's Innovators)."

She also enjoys her time at Lucent Technologies more than her time in academia because it does not seem like such a struggle to get what you want or need in the way of equipment or expense. As long as you do a good job and keep doing a good job the company will provide you with whatever you need. Where as with academia you have to propose grants and get approved before you can continue on your research.

The Mathematics that is involved with Margaret Wright's work is very complex. I was only able to summarize a few key terms about the mathematics involved in finding the best location for a wireless communications tower.

It does not sound like it would be that difficult to find the best location for a wireless communications tower, but there are several underlying mathematical problems that first must be solved.

The first step to solving the problem is modelling. This is when you create a model that is as closely related to the real-world situation as possible. You have to use a model when it is too expensive, dangerous, or just impossible to use the real-world. Even if the model is highly nonlinear it is formed to a linear program because until recently nonlinear methods were not available for large scaled problems. There are many arbitrary decisions that are made which do not affect the accuracy of the model, but are crucial to whether or not an optimization algorithm can be applied to the model. The most general optimization problem is to minimize a scalar function of its independent variables that are subject to restrictions and constraints.

The idea is that you have a smooth function, which produces a smooth curve on a graph. You want to be able to fit your model as closely as possible to the curve with the least amount of discontinuities as possible. Transformation takes place in order to be able to do this. In transformation, you simplify or eliminate constraints. It is very time consuming to pick which constraint is more necessary than others. Also, if you get rid of or over simplify the wrong constraints you may end up with a problem that does not represent what you are trying to find.

They use a term called scaling which is a vague term used to discuss the large range of numerical difficulties that are known to exist in the universe but can not be described in precise general terms. So scaling by variable transformation converts the variables from units that reflect the physical nature to units that have desirable properties during the minimization property.

The aspects in nature that have to be looked at are the oxygen, carbon, hydrogen, and nitrogen balance. With each aspect comes another equation and so your set becomes overdetermined because you have too many variables and not enough unknowns. So the equation gets minimized so that it will best fit the curve it was provided with and that is where the wireless communications tower should be placed. Now that is not always the case because once the location is found, it still has to be bought, and the people in that area have to agree that it can be built there. At least the mathematics of the problem has been solved.


"Margaret H Wright: Distinguished Member of Technical Staff AT&T Bell Laboratories", This source had a lot of information and I used it throughout the paper.

"Today's Innovators", This source was good because of the direct quote from Margaret Wright.

"Margaret H. Wright, The Mathematics of Optimization", This was helpful but did not have much information.

Gill, Phillip E.; Murray, Walter; Saunders, Michael A.; Wright, Margaret H. Aspects of mathematical modelling related to optimization. Applied Mathematical Modelling 5 (1981), no.2, 71--83. The math was based on this paper.

Agnes Scot SiteThis source wasn't very helpful.


Her academic advisor for her PhD was Gene H. Golub

Margaret must think very highly of Phillip Gill and Walter Murray's opinion. She has co-authored two books and many papers with them.