L32_Quadratic Programming - Modified Simplex algorithm

L32_Quadratic Programming - Modified Simplex algorithm -...

Info iconThis preview shows pages 1–11. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: QUADRATIC PROGRAMMING Quadratic Programming Quadratic Programming A quadratic programming problem is a non-linear programming problem of the form DX X X c z T + = → Maximize Subject to → → ≥ ≤ , X b X A Here [ ] n m n c c c c b b b b x x x X . . . , . . , . . 2 1 2 1 2 1 = = = → → = mn m m n n a a a a a a a a a A . . . . . . . . 2 1 2 22 21 1 12 11 = nn n n n n d d d d d d d d d D . . . . . . . . 2 1 2 22 21 1 12 11 We also assume that the n × n matrix D is symmetric and negative-definite. This means that the objective function is strictly concave . Since the constraints are linear, the feasible region is a convex set. In scalar notation, the quadratic programming problem reads: Maximize 2 1 1 1 2 n n j j jj j ij i j j j i j n z c x d x d xx = = ≤ < ≤ = + + ∑ ∑ ∑ ∑ Subject to , . . ,. , . . . . . . . . . . . 2 1 2 2 1 1 2 2 2 22 1 21 1 1 2 12 1 11 ≥ ≤ + + + ≤ + + + ≤ + + + n m n m n m m n n n n x x x b x a x a x a b x a x a x a b x a x a x a The solution to this problem is based on the KKT conditions. Since the objective function is strictly concave and the solution space is convex, the KKT conditions are also sufficient for optimum. Since there are m+n constraints, we have m+n Lagrange multipliers; the first m of them are denoted by λ 1 , λ 2 , …, λ m ; and the last n of them are denoted by μ 1 , μ 2 , …, μ n . The KKT (necessary) conditions are: , . . . , , , , . . . , , . 1 2 1 2 1 ≥ n m μ μ μ λ λ λ n j a x d c j n i ij i n i i ij j , . . . , 2 , 1 2 . 2 1 1 = = +- + ∑ ∑ = = μ λ (continued …) 1 3. 0, 1,2,. . ., 0, 1,2,. . ., n i ij j i j j j a x b i m x j n λ μ = - = = = = ∑ 1 4. , 1,2,. . ., 0, 1,2,. . ., n ij j i j j a x b i m x j n = ≤ = ≥ = ∑ Denoting the (non-ve) slack variable for the ith constraint i n j j ij b x a ≤ ∑ = 1 by S i , the 3 rd condition(s) can be written in an equivalent form as: n j x m i S j j i i , . . . , 2 , 1 , , . . . , 2 , 1 , . 3 = = = = μ λ Referred to as " Restricted Basis " conditions. Also condition(s) (2) can be rewritten as: 1 1 2. 2 1,2,. . ., n n ij i i ij j j i i...
View Full Document

This note was uploaded on 02/23/2011 for the course MATH 112 taught by Professor Ritadubey during the Spring '11 term at Amity University.

Page1 / 38

L32_Quadratic Programming - Modified Simplex algorithm -...

This preview shows document pages 1 - 11. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online