Linear Programming for the   TI-83 Plus/TI-84

Content:  This document covers use of the inequality application, row operations, and simplex programs,
on the TI-83 Plus calculator.
 
INDEX:

To facilitate lookup, the instructions are divided into the following categories:

         I.  Basic Information - types of row operations, improving efficiency of row operations, advantages
                  of inequality application,  discussion of whether to use a simplex program, downloading a simplex
                  program, entering a simplex program manually,
        II.   Linear Inequalities - Linear programming without the inequality App, Linear programming with the
             Inequality APP,
       III.  Simplex Method -Simplex method by hand
       IV. Simplex method using calculator programs - Simplex method using negative slack variables, Simplex
            method using positive slack variables, addition to the positive slack variable program that allows that
            program to use negative slack variables,

RELEASE DATE:  9/30/09         DATE LAST REVISED:  10/18/09

 © 2003 Frank Kizer    NOTE:  See copy restrictions and printing hints at the end of this document.  

 I.  Basic Information:
      1. Updating or not Updating a Matrix with a Row Operation:
           You can choose to only show the result of a row operation or  to show the results and update
            the matrix.  If you want to store the result in the matrix, you must tell the calculator to store
            the matrix.  For example, to swap rows 1 and 2 in matrix [A], you would have rowSwap([A], 1,2).
            On the other hand, if you want to store the result in matrix [A], you must have
             rowSwap([A], 1,2)-->[A].  The symbol --> is made by pressing STO.
      There are four different types of row operations available on the TI-83 Plus as listed below:
           a)  Swap two rows with the snytax rowSwap(matrix name, row number, row number).  For example,
                 if you want to swap rows 1 and 2 in matrix [A] the display would be as rowSwap([A], 1,2).  You can
                 either close the parentheses or not whichever you choose.   Do not type an "A" and enclose it in
                 brackets for the matrix type.  You must press 2ND, MATRIX, select the matrix, and press ENTER.
                If you want to store the information in matrix [A], you must have rowSwap([A], 1,2)-->[A].
          b)  Add two rows with the syntax row+(matrix name, row number, row number).  For example,
                if you want to add row 1 to 2 in matrix [A] the display would be as row+([A], 1,2).  You can
               either close the parentheses or not whichever you choose.   Do not type an "A" and enclose it in
               brackets for the matrix type.  You must press 2ND, MATRIX, select the matrix, and press ENTER.
               If you want to store the information in matrix [A], you must have row+([A], 1,2)-->[A].
         c)  Multiply a row by a number  with the syntax *row(multiplier, matrix name, row number).  For example,
               if you want to multiply row 1 by - 2 in matrix [A] the display would be as *row(-2,[A], 1).  You can
               either close the parentheses or not whichever you choose.   Do not type an "A" and enclose it in
               brackets for the matrix type.  You must press 2ND, MATRIX, select the matrix, and press ENTER.
               If you want to store the information in matrix [A], you must have *row(-2,[A], 1)-->[A].
        d)  Multiply a row by a number  and add it to another row with  the snytax *row+(multiplier, matrix name,
               row number,  row number).  For example,  if you want to multiply row 1 by - 2 in matrix [A] and add it
               to row 2 in matrix [A], the display would be as *row+(-2,[A], 1,2).  You can either close the parentheses
              or not whichever you choose.   Do not type an "A" and enclose it in brackets for the matrix type.  You must
              press 2ND, MATRIX, select the matrix, and press ENTER.  If you want to store the information in matrix [A],
              you must have *row+(-2,[A], 1,2)-->[A].

2.  Improving Efficiency of Row Operations:
         If you have read the simplex method using row operations that follows farther down the page, you will notice
         I only use the *row+( operation in that procedure.  That can be used because it is possible to use that operation
         to do the addition  of two rows and the multiplication of a row.  If that is done, then all one has to do is press
         2ND, ENTRY to display the previous *row+( operation and edit at most three numbers.  I will give two examples
          that will demonstrate that procedure.
        a)  Suppose I want to add row 1 to row 2.  Instead of using row+([A]1,2), use *row+(1,[A], 1,2).
        b)  Suppose you want to multiply row 1 by 1/2.  Instead of using *row(1/2, [A],1), use
              *row+(-1/2,[A],1,1).  Of course, to multiply by 1/3, one would use *row+(-2/3,[A],1,1).
           The obvious advantage of this method is that instead of going to the matrix list to get operations, one
           need only press 2ND, ENTRY and edit the numbers in the previous row operation. 

3. Advantages of Inequality Application:
     The inequality application has several advantages over the method without the application.  I have
     listed the major ones below.
     a)  Probably the most important  advantage is that X-variables are available, so that one can graph lines
           such as x≤3. 
     b)  All shading except that for the feasible region can be eliminated.  This eliminates the confusion of
           determining which regions are applicable to all constraints. 
    c)  The the corner points can be determined without the calculator CALC function.
    d)  The objective function can be evaluated by the calculator and stored in a special list, rather than
           having the student do the evaluation manually.  I am aware, however, that some students may
           not have enough experience with lists to make this worthwhile.

4)  Whether to Use a Simplex Program:  
       Once you get the proper simplex program installed in your calculator, all you need to do is set up the
       proper initial tableau to solve linear programming problems.  There are, however, a few obstacles to
       getting the proper program installed.  The major ones are discussed below.  I should first add that
       everyone who studies linear programming does not need a simplex program.  If you're only going to
       do a few row operations, or maybe one or two simple programs, probably one of the other methods
       described here will be entirely sufficient.
       a)  Choosing the Correct Program:
              First one needs to know what format the text, teacher, or other source will required.  Will you be
              solving only standard maximum problems or will you be solving those as well as standard
              minimum and mixed constraints?  Further, will you require a program that can handle positive
               slack variables, negative slack variables, or both?  These are things that ideally should be settled
               before you install a program.
       b)  Getting the program into your calculator:
              1)  Downloading the Program:  A simplex program can be downloaded from the Internet and
                     installed in a calculator if the file has a .8XP extension.  To install a file with a .8XP file one needs
                     the TI Connect program installed on a computer and either a USB or "Silver" cable depending on
                     which model calculator you have.
               2)   Entering the Program Manually:  The program can be entered by hand if the program is in text
                      format.  When entering by hand, one needs to know where to find the various I/O (input/output)
                      and Control commands.  For a person with some knowledge of where to find these commands,
                      somewhere around twenty to thirty minutes will be required to enter an average-length simplex
                      program. 

II. Solving Linear Programming Problems Graphically:

   1) Graphical Linear Programming without the Inequality Application:  .
       Find the maximum of the objective function z=2x +5y, subject to the following constraints:
            3x+2y≤6  (Eq 1a)
            -x+2y≤4   (Eq 2a)
             x≥0, y≥0
             a)  First put the two-variable equations in slope-intercept form.
                   y≤-3/2x+3  (Eq 1b)
                   y≤1/2x+2   (Eq 2b)
                   (Remember  that you may need to change the direction of the inequality sign if you have to
               multiply or divide by -1 during the rearranging of the equation.)
              b)  Enter the right side of those equations opposite Y1 and Y2 respectively. Opposite the equal sign
                    for Y3 enter 0 (zero).
              c)  Set the WINDOW at Xmin = -1, Ymin=-1, Ymax= the largest value of "b" plus a few units, say 5 in
                    this problem.
                    Xmax is a bit more difficult to anticipate.  If there is an equality with a negative slope, I usually
                    make Xmax = 4/3*b/m, round it off to the next largest whole number and add a few units. You can
                    enter the arithmetic opposite Xmax. You might want to press GRAPH and see if all of the
                    corner points of the bounded region are on the screen.
              Now we will enter the inequality signs:
           a)  Shading of the graph is determined by the symbol to the left of the "Y=" entry.  Using the left
                arrow, move the cursor all the way to the left of the Y= symbol.
           b)  Pressing ENTER in that position will display different symbols.  For < or <, press ENTER
                until the upright triangle is displayed.  For > or >, press ENTER until the upside down
                triangle is displayed.
           c) After you have the correct symbol displayed, press ENTER to graph the inequality.
           Finding  the x- and y-values of the corner points. 
                You may find it easier to locate the corner points if you change the inequalities back to equalities
                 and press GRAPH to graph only the lines.  Write down the values of each corner point.  They
                 will be used to evaluate the objective function
                a) Press 2^nd,  CALC.
            b) Press 5 to select intersect.
            c)  To get the corner point at the intersection of the two graphs, move the cursor  a little away from
                  the intersection and press ENTER to mark the first graph line and move the cursor to the next
                  graph line.
            d)  Move the cursor to a little away from the intersection and press ENTER to mark the second graph
                 line and set the cursor to guess the intersection.
            e) Move the cursor approximately to the intersection and press ENTER.
            f) The coordinates for the point of intersection, x=.5 and y=2.25, will appear at the bottom of the screen.
            g)  To obtain the value where the graph line intersects the x-axis, repeat the procedure
                 for the intersection of two lines except that for the second curve you will need to insure
                 that the cursor is on the x-axis.  If it is not, press the down arrow. The answer is x=2, y=0.
            h)  The corner point where the graph intersects the y-axis is obtained by pressing 2ND, CALC,
                 pressing ENTER to select Value,  entering 0 (zero) opposite X, and then pressing ENTER.
          Having the calculator evaluate the objective function.
               a)  First you need to enter the proper expressions.  From the home screen, press .5, STO, X, ALPHA, :
                (decimal point key), 2.25, STO, ALPHA, Y, ALPHA, :, 2, X, +, 5, ALPHA, Y.  You should now
                have the following on the home screen:  .5→X:2.25→Y:2X+5Y
               b)  Press ENTER and the corner point will be evaluated at 12.25.
               c)  To evaluate additional points, press 2nd, ENTER and the above expression will be displayed.
                     Enter the new corner point and press ENTER.  Repeat this as many times as needed.

 2)  Graphical Linear Programming with the Inequalities Application:
         Find the maximum of the objective function z=2x +5y, subject to the following constraints:
            3x+2y≤6  (Eq 1a)
            -x+2y≤4   (Eq 2a)
             x≥0, y≥0
           a)  First put the two-variable equations in slope-intercept form.
                 y≤-3/2x+3  (Eq 1b)
                 y≤1/2x+2   (Eq 2b)
           b)  Enter the right side of those equations opposite Y1 and Y2 respectively and enter 0 opposite Y3.
           c)  Set the WINDOW at Xmin = -1, Ymin=-1, Ymax= the largest value of "b" plus a few units, say 5 in
                 this problem.
                Xmax is a bit more difficult to anticipate.  If there is an equality with a negative slope, I usually
                 make Xmax = 4/3*b/m, round it off to the next largest whole number and add a few units. You can
                  enter the arithmetic opposite Xmax. You might want to press GRAPH and see if all of the
                  corner points of the bounded region are on the screen.
           Now we will enter the inequality signs.
         a)  Move the cursor to the sign (either equal or inequality) after Y1.  If the inequality symbols do not
                   appear at the bottom of the screen, you will need to start the Inequality App.  Do that by pressing
                   APPS, move the cursor down to Inequal,  or Inequalz for the international version, and press
                   ENTER, ENTER.  The Y= editor screen should be displayed. 
         b)  Place the cursor on the equal sign opposite Y1 and press ALPHA, F3 (ZOOM).  The equal sign
              should have been replaced by the inequality ≤.
             c)  Do the same for Y2; then opposite Y3, press ALPHA, F5 (GRAPH).  The symbol ≥ should have
                   replaced the equal sign before the 0. 
             d)  Now, move the cursor up to the "X" in the upper left corner and press ENTER.  With the cursor on the
                   equal sign opposite X1, press ALPHA, F5 (GRAPH) to enter ≥;  then enter a zero after that symbol.
             e)  Press GRAPH  to draw and shade the graphs.  
             f)  Press ALPHA, F1, 1 to define the feasible region.
             If you only want to graph, you may stop here.
             At this time we will find the x- and y-values of the corner points.  We will use a method to have the
             calculator determine the corresponding values of the objective function.  If you prefer to determine
             those values manually, just record them without pressing STO so that you can substitute them into
             the objective function.
             a)  Press ALPHA, F3 (ZOOM) and  x=.5, y=2.25 will be displayed.  Record these values if you choose to evaluate the objective
                   function by hand.  Otherwise, press STO, ENTER.
              b) Press the left arrow to move to the point x=0, y=2 and press STO, ENTER. 
              c)  Press the right arrow; then then down arrow to read the last point, x=2, y=0. 
             Having the calculator evaluate the objective function.
              a)  Press STAT, ENTER, to bring up the lists with the stored data.
              b)  First we will name the list after list INEQY.  We will name it OBJ.  Move the cursor to the name
                    block at the top of that list.  Press 2ND, ALPHA, O, B, J, ENTER.
              Now, we will define the list OBJ.
              c)  With the curosor on the list name, press ALPHA, " (the + key), 3, x(multiply), 2ND, LIST and scroll down to
                    INEQX and press ENTER.
             d)  Press +, 2, x (multiply), 2ND, LIST, scroll to INEQY and press ENTER.  Press ALPHA, ".  You should now have
                   "3*└INEQX+2*└INEQY" at the bottom left of the screen after "OBJ=".  Press ENTER and the values
                   of the objective function for each set of coordinates will be displayed in the OBJ list. 

III. Simplex Method:
       1.  Simplex Using Row Operations with the TI-83Plus or TI-84:
                 First of all, I suggest that you enter your first tableau in matrix [B], and then use STO to transfer it to matrix [A].
             That way you can work on matrix [A]  and if you make a mistake you won't need to re-enter the original matrix. 
             You can Just  just transfer the information from [B] to [A] again.   Transfer [A] to [B] as follows:   Press 2ND, MATRIX,
              ENTER, STO, 2ND, MATRIX, 2. 
       Let's say that you have this set of equations:
        Objective Function:
         z=24x +36y
       Constraints:
         x +2y ≤ 14
        3x +4y ≤36
        x≥ 0, y≥0
      So that,  when you add slack variables, you the following as the  first tableau:
        x      y     S₁    S₂   
       1       2     1       0| 14
       3       4     0       1|36
     -24   -36  0       0|0

    Now,  for the pivot point we first pick the column with the most negative number in the objective row.  That's -36,
    so we pick column 2.  Now, we want to select the proper row.  Divide 2 into 14and 4 into 36 and select the smaller
    result.  The quotient for 14/2  is smaller, so 2 is the pivot point. You want to multiply row 1 by -2 and add it to row 2. 
     I'll use "shorthand" such as -2R1 + R2=>R2, to indicate this type of operation. 
     a)  So, enter the matrix in matrix [A] and store it in matrix [B] as indicated above; then press 2nd, QUIT to go
           to the home screen.
    b)  From the home screen, press 2nd, MATRIX; scroll over to MATH; scroll down to item F, and press ENTER.  You
         should now have *row+( on your home screen.
   c)  Now you want to enter the multiplier, letter of the matrix, row number multiplied by, row number to replace in
        that order.  Do this:  Enter -2, press the comma button, press 2nd, MATRIX, ENTER.  You should now
        have *row+(-2, [A] on your screen.  DO NOT enter [A] with the brackets and the letter A but get this from the
        matrix list.
  d)  Now press the comma button, enter 1;  press the comma button and enter 2.  You should now have
        *row+(-2,[A],1,2 on your screen.
  e)  Press STO, 2nd, MATRIX, ENTER. 
  f)  You should now have *row+(-2,[A], 1,2)->[A] on your screen.  DO NOT enter [A] with the brackets and the
        letter A.  (Note carefully that if you don't add the STO, [A];  then matrix A will not be changed in your calculator.  
       Also notice that you can either close the parentheses after the last row number or leave them open.)  Press ENTER
 g)  To get a zero in column 1 of the 3rd row, press 2ND, ENTRY (the ENTER key)  to display the first entry again. 
       Now edit the display so that you have this:  *row+(18,[A],1,3)->[A].  Press ENTER
 h)  You now have this matrix:
     [1       2     1       0| 14
      1      0     -2       1|8           -2R₁+R₂==>R₂
     -6       0     18       0|252]    18R₁ +R₃==>R3
     Now we want to have zeros in the first and third position of column 1.
      i)  Press 2ND, ENTRY to display the expression and edit it to get this: *row+(-1,[A], 2,1)->[A].
      j)  Now, to get a zero in column 1, row 3, press 2ND, ENTRY and edit the expression to get this:
     *row+(6,[A], 2,3)->[A].
     You now have this matrix:
      [0       2     3       -1| 6   -R₂+R₁==>R₁
       1      0     -2       1|8
       0       0     6       0|300]  6R₂+R₃==>R₃
    k)  Finally, we need to get a 1 in row 1, column 2.  We could compose a new statement  using the *row( expression,
         but let's just save several steps by pressing 2ND, ENTRY and editing the expression to get this: 
         *row+(-1/2,[A],1,1)->[A].
       The problem is now solved with the following matrix.
       [0       1     3/2    -1/2| 3         -1/2R₁+R₁==>R₁
        1      0     -2              1|8
        0       0     6              0|300]
        x=8, y=3, z=300

       NOTE:    If you want to add two rows; then the syntax is this: row+(matrix letter, row to be added,
             row to be replaced)->[A]   To swap rows, use this syntax:  rowSwap(matrix letter, first row, second row)->[A]
 

THE REST OF THIS SECTION IS UNDER CONSTRUCTION BUT CAN BE USED AS A GUIDE TO THE PROGRAMS
THAT I HAVE WRITTEN.                

2.  Simplex  with the TI-83Plus or TI-84 Using Calculator Programs:
      General:  Calculator programs for doing simplex and quite different in the details of  their approach, but the
      differences in those that I have seen boil down to whether they will do standard maximization, standard
      minimization,  mixed, more than of these, or all of them.  There are many program on the Internet that can be
      found by a quick search with a good search engine.  I will list my programs and describe what they will do and
      where they can be found.
   a)  Simplex Program Using Negative Slack Variables (tisimplex_neg):   This stand-alone program is for those who are familiar with the simplex method that uses negative slack variables when doing problems with mixed constraints or minimization.  You must enter the first tableau in matrix [A] with the proper slack variables and with the proper signs for the indicator row (objective function.)  The program then manipulates rows to give a first feasible solution and displays the solution in decimal form.  The solution may be displayed in fractional form, if appropriate, by pressing ENTER.  When you are finished with the answer, press ENTER because the program is STILL RUNNING in PAUSE Mode to permit scrolling the matrix.   This program can be found by clicking on the link in the title at the beginning
of this discussion. 
  b) Simplex Program Using Positive Slack Variables (tisimplex_pos):   This program is for those who are familiar with the simplex method that uses POSITIVE slack variables when doing problems with mixed constraints or minimization.  You must enter the first tableau in matrix [A] with the proper slack variables and with the proper signs for the indicator row (objective function.)  The program then manipulates rows to give a first feasible solution and displays the solution in decimal form.  The solution may be displayed in fractional form, if appropriate, by pressing ENTER.  When you are finished with the answer, press ENTER because the program is STILL RUNNING in PAUSE Mode to permit scrolling the matrix. I am not the originator of the ideas in this program.  This program can be found by clicking on the link in the title at the beginning  of this discussion. 
c)  Simplex Program Extension using Negative Slack Variables (tisimplex_ext):   This is a menu driven front-end extension to the above program for positive slack variables, that permits use of negative slack variables with that program.  You must have the program tisimplex_pos entered in your calculator in order for this program to give you anything other than the initial tableau for minimizations or mixed constraints.   This program sets up the first feasible tableau for mixed inequalities and standard minimization problems.  The program tisimplex_pos and then solves that tableau  for the final optimized tableau.  Appropriate negative slack variables must entered for mixed, or minimization problems.  Of course, you can solve standard maximization problems by entering the first tableau with appropriate positive slack variables and choosing the appropriate item from the menu.  This program can be found by clicking on the link in the title at the beginning  of this discussion. 
==================================

Copy Restrictions:  You may make single copies of this document for your own personal use and for the use of other students, but inclusion in another document, publication or any use for profit requires my permission.  Teachers may make multiple copies of this document for their students if they first get permission from me.  Merely send me an email (Just click on Webmaster in the navigation bar.) with a one-sentence explanation of what you’re using the document for.  I’ll give you permission in a timely manner. 

Making it Better:  I would be grateful if you would report any errors or suggestions for improvements to me.  Just click "E-mail Webmaster," site the item number, and tell me your suggested change.

Printing Hint:  Most browsers will send both the navigation bar and the text to the printer, and, as a result, will cut off the right edge of this document if it the file is printed directly.  To prevent this, highlight the instructions portion only (not the navigation panel) and check "Selection" on the Print dialog box; then click "Apply."  This will eliminate the navigation panel and get all of the instructions on the printed pages.