Quadrex Algorithm for Negative Definite Quadratic Programming Models

Mark Ivan P. Arcillas *

Department of Applied Mathematics, College of Science and Mathematics, University of Science and Technology of Southern Philippines, Cagayan de Oro City 9000, Philippines.

Elmer C. Castillano

Department of Applied Mathematics, College of Science and Mathematics, University of Science and Technology of Southern Philippines, Cagayan de Oro City 9000, Philippines.

*Author to whom correspondence should be addressed.


Abstract

In this paper, a quadrex algorithm for quadratic programming problems is introduced (n = 2) under linear and quadratic constraints. The quadrex algorithm considers on the behavior of the quadratic function near the origin or a translate of the origin, performs a series of translations
and orthogonal rotations to obtain the optimal solution of the objective function as well as taking considerations on the constraints of the problem. The method works provided that the eigenvalues of the matrix on quadratic form of the objective function is strictly negative, that is,
Q is negative-definite. The quadrex algorithm is a parallel counterpart of the simplex algorithm for linear programming models.

Keywords: Simplex, quadratic, quadratic programming, quadrex, NP-hard, negative denite.


How to Cite

Arcillas, Mark Ivan P., and Elmer C. Castillano. 2022. “Quadrex Algorithm for Negative Definite Quadratic Programming Models”. Journal of Advances in Mathematics and Computer Science 37 (6):57-65. https://doi.org/10.9734/jamcs/2022/v37i630461.

Downloads

Download data is not yet available.