This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Math 171A: Mathematical Programming Instructor: Philip E. Gill Winter Quarter 2009 Homework Assignment #2 Due Friday January 23, 2009 The starred exercises require the use of Matlab . Remember that it is necessary to do all the Matlab assignments to obtain credit for the class. Exercise 2.1. (a) Write the constraints x 1 + x 2 4, x 1 + 3 x 2 6, 6 x 1 x 2 18, 3 x 2 6 and x 1  1 in matrix form Ax b . After rearranging some of the constraints we find that A = 1 1 1 3 6 1 1 1 1 and b = 4 6 18 3 6 1 . (b) Draw the graphical representation of the feasible region defined by the constraints of part (a) . Mark the infeasible side of the halfspace defined by each constraint. Below is the graph with the feasible region. The vertices are A = ( 1 , 5) , B = (1 , 3) , C = (3 . 5 , 3) , D = (4 , 6) , E = ( 1 , 6) 1 2 3 4 5 61 1 2 3 4 5 feasible region A E B C D x 1 x 2 2 Mathematics 171A (c) Consider the linear function ( x ) = c T x where c T = (2 3). On your graph from part (a) , show the halfspace of descent directions emanating from the point x = (2 , 4). The linear function ( x ) = 8 = c T x is shown in the graph below. The halfspace of descent directions emanating from x is the space above this line. 1 2 3 4 5 61 1 2 3 4 5 feasible region A E B C D x 1 x 2 c x (d) Use the graphical method to minimize 2 x 1 3 x 2 subject to the constraints of part (a) . Give the optimal values of x 1 , x 2 and the minimum value of the objective function. If the objective level curve is shifted along the negative normal of ( x ) we find that the solution x * is located at E = ( 1 , 6) , which lies at the intersection of the hyperplanes defining the constraints x 1  1 and x 2 6 . The minimum value of the objective is 2( 1) 3(6) = 20 . (e) Compute the residual vector r ( x ) for the constraints at the point x and find the con straints whose residuals would decrease after a positive step along the direction p = 1 2 emanating from x ....
View Full
Document
 Winter '08
 staff
 Math, matlab

Click to edit the document details