Abstract
Mathematical programming computer systems using the product form in the inverse (PFI) must periodically resort to a reinversion with the current basis in order to reduce the amount of work to be done in the succeeding iterations.
In this paper, we show the consequences of column, pivot selection and sequence upon the transformation vector (ETA) density and give an algorithm which will tend to minimize eta density and work done per iteration.
The algorithm has been implemented and tested as a replacement for the previous inversion algorithm on the OPTIMA system for the CDC 6000 computers and on the MPS/III mathematical programming system for the IBM 360 computer. A comparative performance table is given.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
H.M. Markowitz, “The elimination form of the inverse and its application to linear programming,”Management Science 3, No. 3 (1957) 255–269.
D.V. Steward, “On an approach to techniques for the analysis of the structure of large systems of equations,”SIAM Review 4, No. 4 (1962) 321–342.
P. Wolfe and Leola Cutler, “Experiments in linear programming,”Recent advances in mathematical programming (McGraw-Hill Book Co. Inc., New York, 1963) pp. 177–200.
G. Zoutendijk,Methods of feasible directions (Elsevier Publishing Co., Amsterdam) pp. 52–55.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Hellerman, E., Rarick, D. Reinversion with the preassigned pivot procedure. Mathematical Programming 1, 195–216 (1971). https://doi.org/10.1007/BF01584086
Received:
Revised:
Issue date:
DOI: https://doi.org/10.1007/BF01584086
