lecture9 - infeasible yes no e62: lecture 9 13/10/10...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
e62: lecture 9 13/10/10 Simplex Algorithm 1 c e62: lecture 9 13/10/10 Simplex Algorithm 2 simplex algorithm problem data initial vertex ( A, b, c ) v today’s version assumes existence of an optimal solution max cx s . t . Ax = b x 0 standard form LP x * optimal vertex e62: lecture 9 13/10/10 Vertices Vertex = intersection of N linearly independent constraint boundaries M come from Ax = b N ! M are nonnegativity constraints At least N ! M components are zero N ! M nonbasic variables M basic variables Example N =5, M =3, B = {1, 4, 5} Solving for a vertex given B 3 max cx s . t . Ax = b x 0 } } M constraints N constraints B B 0 1.5 3 4.5 6 1 2 3 4 5 Ax = b x B =0 A B x B = b x B =0 x B = A - 1 B b x B =0 e62: lecture 9 13/10/10 Edges Def. adjacent vertices: share N-1 constraint boundaries An edge is a part of the line that satisFes these constraints Swap one constraint to switch between adjacent vertices In a symmetric form LP, swap one nonbasic variable 4 0 1.5 3 4.5 6 1 2 3 4 5 0 1.5 3 4.5 6 1 2 3 4 5 adjacent
Background image of page 1

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

View Full DocumentRight Arrow Icon
e62: lecture 9 13/10/10 Traversing an Edge Example N =5, M =3, B = {1, 4, 5} 5 0 1.5 3 4.5 6 1 2 3 4 5 feasible to increase ? pick nonbasic variable increase until
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: infeasible yes no e62: lecture 9 13/10/10 Reduced Profts Def. reduced proFt = objective value increase per unit nonbasic variable increase 6 1.5 3 4.5 6 1 2 3 4 5 x- x + cx- cx + c reduced proft = c 2 e62: lecture 9 13/10/10 Algorithm and Convergence Termination At an optimal solution, no positive reduced profts Finite time convergence There are a fnite number o vertices 7 1.5 3 4.5 6 1 2 3 4 5 compute reduced costs traverse maximizing edge maximum positive ? yes no e62: lecture 9 13/10/10 Degeneracy and Cycling Def. degeneracy = when there is a zero-valued basic variable Example where N =5, M =3 (3 basic variables) Intersection o N+1 constraint boundaries Def. cycling = keep swapping basic variables without changing solution 8 1.5 3 4.5 6 1 2 3 4 5...
View Full Document

Page1 / 2

lecture9 - infeasible yes no e62: lecture 9 13/10/10...

This preview shows document pages 1 - 2. Sign up to view the full document.

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