Item description for Stochastic Linear Programming Algorithms: A Comparison Based on a Model Management System (Optimization Theory & Applications Series) by Janos Mayer...
A computationally oriented comparison of solution algorithms for two stage and jointly chance constrained stochastic linear programming problems, this is the first book to present comparative computational results with several major stochastic programming solution approaches. The following methods are considered: regularized decomposition, stochastic decomposition and successive discrete approximation methods for two stage problems; cutting plane methods, and a reduced gradient method for jointly chance constrained problems. The first part of the book introduces the algorithms, including a unified approach to decomposition methods and their regularized counterparts. The second part addresses computer implementation of the methods, describes a testing environment based on a model management system, and presents comparative computational results with the various algorithms. Emphasis is on the computational behavior of the algorithms.
Promise Angels is dedicated to bringing you great books at great prices. Whether you read for entertainment, to learn, or for literacy - you will find what you want at promiseangels.com!
Studio: CRC Press
Est. Packaging Dimensions: Length: 0.75" Width: 7.25" Height: 10.25" Weight: 1.15 lbs.
Release Date Feb 25, 1998
ISBN 9056991442 ISBN13 9789056991449
Reviews - What do customers think about Stochastic Linear Programming Algorithms: A Comparison Based on a Model Management System (Optimization Theory & Applications Series)?
Interesting for researchers; not so much for students. Apr 14, 2000
This not a bad book, and I didn't regret buying it. However, if you are interested in an introduction to stochastic linear programming, look for other books (e.g. Kall/Wallace or Birge/Louveaux). Indeed, the title does not convey the true nature of the book. Its subtitle (A comparison based on a model management system) better clarifies its purpose: it is a short research monograph where different solution algorithms are compared. If you want, it is a sort of high-level thesis work, from which a book has been derived.
Computational tests are a quite significant part of the book, whereas the introduction to the field is quite brief (and limited in scope; for instance only two-stage problems with recourse are described).
The theoretical part aims at being rigorous and self-contained (since it starts from the basics of convex optimization), but it is too brief and hard to read for a student. It can be used as a reference for the experienced reader, provided she tracks all the required references in the literature. Furthermore, it may be interesting for people searching for some work on chance constrained problems, since the author is an active researcher in this area.
As to the computational testing, it is worth noting that the author describes how to generate problem instances to test different algorithms; this is good for people looking for benchmarks. On the other hand, many readers will not find the description of the codes developed by the author much useful.
On the whole, while it may be a useful reading for a researcher, this is more a long survey article than a true book. In view of this, a cheaper edition would have been better.