Computational Experiments on Large-Scale Optimization with the Decomposition Principle.
Abstract
In this paper, the solution of large optimization problems for regional planning are traced out by a decomposition program that is based on the use of standard LP programs. The underlying method is the decomposition principle of Dantzig and Wolfe. The concept is tested by solving an optimization problem with about 1250 columns and 900 rows. Furthermore, it is investigated to what extent the efficiency of the solution procedure can be influenced by an appropriate choice of starting solutions or specific matrix divisions.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1978
- Accession Number
- ADA063755
Entities
People
- Gerhard Schiefer
Organizations
- Stanford University