This preview shows pages 1–7. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: IE 495 Lecture 25 November 30, 2000 Reading for This Lecture Primary Bazaraa, Sherali, and Sheti, Chapter 2. Chvatal, Chapters 6 and 7. Linear Programming Introduction Consider again the system Ax = b, A R m n , b R m . In this problem, there are either no solutions one solution infinitely many solutions (if n > m) The problem of linear programming is min c T x s.t. Ax = b x 0 The Simplex Algorithm Note that x B = B1 b  B1 Nx N Hence, c T x = c B T x B + c N T x N = c B T B1 b + ( c N T c B T B1 N)x N So if c N T c B T B1 N , we have found the optimal solution (why?). Otherwise, suppose some component of c N T c B T B1 N is negative. Then we raise the value of the corresponding variable as much as possible while maintaining feasibility. Summary of the Simplex Algorithm Simplex algorithm Compute yB = c B T Choose a column of a j of N such ya j < c j Compute Bd = a j Find the largest t such that x B *  td 0 Set the value of x j to t and the values of the basic variables to x B *  td ....
View Full
Document
 Fall '08
 Linderoth
 Operations Research

Click to edit the document details