Solutions to Linear Programming Formulation Examples
(Distributed in class)


  1. Better Products...
  2. Max 30 X1 +  50X2 +  20X3
    st.
    .5x1 + 2x2 + .75x3 <=  40
      x1 +  x2 + .5 x3 <=  40
     2x1 + 5x2 +   2x3 <= 100
      x1 <= .5*(x1 + x2 + x3) simplifies to ( .5x1 - .5x2 -.5x3 <= 0)
      x3 >= .2*(x1 + x2  +x3) simplifies to (-.2x1 - .2x2 +.8x3 >= 0)
      x1, x2, x3 >= 0
    Solver
  3. The Westchester Chamber .....
  4. Max 100,000 XT + 18,000XR + 40,000 XN
    st
    2,000 XT + 300 XR  + 600 XN <= 18,200
     XT <= 10
     XR <= 20
     XN <= 10
     XT >= .1*( XT + XR + XN) simplifies to (.9XT -.1XR -.1XN >=0)
     XR <= .5*(XT + XR + XN)  simplifies to (-.5XT +.5XR -.5XN <=0)
     XT, XR, XN >= 0
    Solver
  5. Shiploading
  6. Xij is the tons of cargo i = (A, B, C) taken in hold j=(f, c, a)

    MAx  25 *(XAf + XAc + XAa) + 30 *(XBf + XBc + XBa) + 20 *(XCf + XCc + XCa) 
    st
    (Hold capacities)
    XAf + XBf + XCf <= 2,000
    XAc + XBc + XCc <= 3,200
    XAa + XBa + XCa <= 1,900

    (Hold volume Capacities in cf)
    60 XAf + 50 XBf + 25 XCf <= 100,000
    60XAc + 50 XBc +25 XCc <=  140,000
    60XAa +50 XBa +25 XCa <=  95,000

    (Available Cargo)
     XAf + XAc + XAa  <= 7,000
     XBf + XBc + XBa <= 6,500
     XCf + XCc + XCa <= 4,000

    (Trim Constraints ) 
    (XAf + XBf + XCf )/2,000 =( XAc + XBc + XCc)/ 3,200
    (XAc + XBc + XCc)/ 3,200 = (XAa + XBa + XCa )/ 1,900

    All  xij >= 0

    Solver
  7. Bootlegger..

  8. Let xij be the fifth of ith whiskey (A,B,C) blended in the jth blend (G,U,O). Profit margin of each of the variable is equal to selling price of the blend less the cost of the whiskey.. For instance for x BG is $6.00 - $ 5.5 = $ .5 per fifth. etc.
    Max   -.5XAU - 1.5 XAO+.5XBG -XBO + 3.1XCG+2.6XCU +1.6XCO
    st:
    Availability of Whiskeys A,B,C
            XAG + XAU + XAO <= 2,800
            XBG + XBU + XBO <= 2,000
            XCG + XCU + XCO <= 2,000
    Blend Specs...Gasp-for-air
            XAG >= .6*(XAG + XBG + XCG)  or  -.4XAG  + .6XBG + . 6CG <= 0
            XBU <= .3*(XAG + XBG + XCG)  or  -.3XAG + .7XBG  - .3XCG <= 0
            XBG + XCG <=.4*(XAG + XBG + XCG)   or .6XBG + .6XCG  - .4XAG <= 0
                    Ulcer Maker:
            XCU <=.5*(XAU + XBU + XCU)     or    -.5XAU  - .5XBU + .5XCU  <= 0
            XAU >=.2*(XAU + XBU + XCU)     or    -.8 XAU +. 2XBU + .2XBU  <= 0
                    Old Black:
            XCO <=.7*(XAO + XBO + xCO)     or   -.7XAO - .7XBO + .3XCO <=0
            all variables >=0
    Solver
  9. Doug Star
  10. Let A, B and C be the quantities of the grains.

    Min             0.45 A + 0.38 B + 0.27 C
    s.t             0.62 A + 0.55 B + 0.36 C >= 8/16
                    0.65 A + 0.10 B + 0.20 C >= 1/16
                    0.03 A + 0.02 B + 0.01 C <= 1/32
                    A, B, C >= 0
    Solver
  11. Edwards manufacturing co.

  12. Let xij be the number of ith component (1,2) purchased from the jth (1,2,3) supplier.
    Min  12 x11 + 13x12 + 14 x13 + 10 x21 + 11x22 + 10x23
    st
    Supplier capacities:
            x11 + x21 <= 600
            x12 + x22 <= 1000
            x13 + x23 <= 800
    Requirement Constraints:
            x11 + x12 + x13 = 1000
            x21 + x22 = x23 = 800
            all variables >= 0
    Solver
  13. Tim Burr company..

  14. Let L be thousands of board feet of lumber produced and P be thousands of square feet of plywood produced.
            Max 45L +  60P
            St
                  L + 2P <= 32     (Availability of spruce0
                 4L + 4P <= 72     (Availability of Douglas Fir)
                  L      >= 5      (Market committment for Lumber)
                       P >= 12     (Market committmant for Plywood)
                  L, P >= 0
    Solver
  15. Waiter Scheduling..

  16. There are 12 scheduling periods running from 11-Noon to 9pm - 10 pm. Let the number of part time employees starting their four hour shifts at the start of period i be denoted by Xi. For instance,  X1 is the number of waiters starting their shift at 11 (period 1) and will be on hand for four consequtive periods (in the table X1 appears in perids 1 through 4), likewise for other periods.  The minimum number of part-time employees needed for each period is computed net of the full time employees available as: total number needed minus 1 if the first full time employee is on minus 1 if the second full-time employee is on for each period. Notice that the last group comes in at 6 pm to work until closing.
     
    Number starting their shift at
    X1 X2 X3 X4 X5 X6 X7 X8
    Period Need 11am 12n 1pm 2pm 3pm 4pm 5pm 6pm
    11 - 12N    (9-1-0)= 8 yes






    12N - 1   (9-1-0)= 8 yes yes





    1 - 2   (9-1-1)= 7 yes yes yes




    2 - 3   (3-1-1)= 1 yes yes yes yes



    3 - 4   (3-0-1)= 2
    yes yes yes yes


    4 - 5   (3-1-1)= 1

    yes yes yes yes

    5 - 6   (6-1-0)= 5


    yes yes yes yes
    6 - 7 (12-1-1)=10



    yes yes yes yes
    7 - 8 (12-1-1)=10




    yes yes yes
    8 - 9   (7-0-1)= 6





    yes yes
    9 -10   (7-0-1)= 6






    yes
      From the table notice that the number of waiters that will be on hand during,  say, perid 3pm -4pm will be X2 + X3 + X4 + X5  which needs to be at least 2 (the number of emplyees needed for that period).. Likewise for all periods.
    Therefore the problem is to:
    Min X1 + X2 + X3  X4 +X5 + X6 + X7 + X8
    st
            X1 >= 8
            X1 + X2 >= 8
            X1 + X2 + X3 >= 7
            X1 + X2 + X3 + X4 >= 1
               + X2 + X3 + X4 + X5 >= 2     
                    + X3 + X4 + X5 + X6 >= 1
                         + X4 + X5 + X6 + X7 >= 5
                           + X5 + X6 + X7 + X8 >= 10
                                   + X6 + X7 + X8 >= 10
                                        + X7 + X8 >= 6
                                             + X8 >= 6
            X1, X2, X3, X4, X5,X6,X7,X8 >=0
    Solver