2011S: Optimization (168)

Instructor: Prof. Matthias Köppe

TA: Robert Hildebrand

Time and Location

Wellman 1,

MWF 0310-0400P

Textbook

Robert J. Vanderbei, Linear programming foundations and extensions, 3rd edition

ISBN-13: 978-0387743875

The book is available electronically via the UC Library at http://dx.doi.org/10.1007/978-0-387-74388-2

(If you're connecting from off-campus, you'll need to use the UC Library VPN.)

You can also order a print copy (soft cover) for $24.95 from SpringerLink.

Topics

We will mostly follow the department syllabus.

We will cover one additional topic not on the department syllabus:

Integer linear optimization.

Grading

10% Class participation*

20% Homework

30% Midterm exams

40% Final exam

There will be two midterm exams. The lower grade of the two midterms

will be dropped.

Homework will be announced on SmartSite. No late homework will be

accepted, unless there is a documented emergency.

* Lectures will be interactive. I will ask questions on material from previous lectures and the required reading, and general questions that require to actively follow the current class.

Homework and announcements will appear on SmartSite. Auditors: If you would like to audit this class, send me your UC Davis id, and I can add you to SmartSite.

Exam Dates

Midterm 1 (in class): Wednesday, Apr 27

Midterm 2 (in class): Friday, May 20

Final exam: as announced on MyUCDavis

Modeling/Computational Homework Execrcises

Homework will include proof exercises, modeling/computational

exercises and programming exercises.

If you don't have an account on the department computers, you need to get a class

account from http://www.math.ucdavis.edu/comp/class-accts

We will be using the ZIB Optimization Suite (SCIP, SoPlex, ZIMPL) in class.

Please read the ZIMPL (optimization modeling language) documentation at:

http://zimpl.zib.de/download/zimpl.pdf

The software is installed on the departmental computers in the directory ~mkoeppe/public/258b

Type

~mkoeppe/public/168/scip

to start the SCIP shell.

You can read ZIMPL models by typing

read MODELFILE.zpl

Use the interactive documentation of SCIP by typing

help

to learn about SCIP.

Please follow the examples in the ZIMPL documentation using SCIP.

You can also install SCIP on another computer. The download page (source and precompiled binaries) is here: scip.zib.de/download.shtml

Precompiled versions for Windows and Linux are available. I don't know whether Mac OS X is supported, but you can try compiling the source code.

Programming Exercises

There will also be programming exercises as part of the homework. They can be completed in a high-level programming

language of your choice.

Additional Reading

The following books are useful additional reading.

  • Jiri Matousek and Bernd Gärtner: Understanding and Using Linear Programming, Springer, 2007
  • Available as an electronic text at http://dx.doi.org/10.1007/978-3-540-30717-4 (If you're connecting from off-campus, you'll need to use the UC Library VPN.)
  • Chapter 4 provides a nice introduction into the polyhedral geometry behind linear optimization and the simplex method.
    • Bradley, Hax, Magnanti: Applied Mathematical Programming
    • A general introduction to mathematical optimization, including integer linear optimization, from an applied point of view. This is a bit dated (1977), but still a good reading on the basic material.
    • A re-typeset version of this 1977 MIT classic is available online as a full text