
274 Introduction to Linear Optimization and Extensions with MATLAB
R
P MP
T
= LBL
T
is obtained where M is a symmetric matrix (for EQPs M = K), P is a
permutation matrix, L is a lower triangular matrix and B is a block diagonal
matrix with block matrices of dimension 1 or 2; see Golub and Van Loan
(1989) for more details.
7.5.5 Inequality Constrained Convex Quadratic Program-
ming
We now consider quadratic programs of the following form
(Q) minimize f(x) =
1
2
x
T
Qx + c
T
x
subject to Ax = b
x ≥ 0
with equality and inequality constraints, i.e., non-negativity restrictions on the
variables. We refer to this class of problem as EIQP. An EIQP can model the
situation ...