PRE FINAL REVIEW EXERCISE The IT Division of a university processes students applications to enter the institution.

During peak period, the processing of the applications is done by a large pool of computer operators, some of whom are permanent and some temporary. A permanent operator can process 24 applications per day, whereas a temporary operator can process 16 per day. On average, the IT Division processes at least 600 applications each day. The IT Division has allocated 45 computer workstations to process the applications. A permanent operator generates about 1 application with errors each day, whereas a temporary operator 2 errors per day. The Director of the IT Division wants to limit applications with errors to 60 per day. A permanent operator is paid RM60 per day, and a temporary operator RM35 per day. The Director wants to determine the number of permanent operators and temporary operators to hire in order to minimize costs.
1.

a. Formulate a Linear Programming model for the problem. b. Solve this model by using graphical method. c. If the temporary operator is paid RM40 per day instead of RM35, what will be your optimal solution? 2. A sawmill in Tannassee produces cherry and oak boards for a large furniture manufacturer. Each month the sawmill must deliver at least 5 tons of wood to the manufacturer. It takes the sawmill 3 days to produce a ton of cherry and 2 days to produce a ton of oak, and the sawmill can allocate 18 days out of a month for this contract. The sawmill can get enough cherry to make 4 tons of wood and enough oak to make 7 tons of wood. The sawmill owner has developed the following linear programming model to determine the number of tons of cherry (x1) and oak (x2) to produce to minimize cost. Minimize Subject to: Z = 3x1 + 6x2 3x1 + 2x2 18 x1 + x2 5 x1 4 x2 7 x1, x2 0 (cost, $) (production time, days) (contract, tons) (chery, tons) (oak, tons) (non-negative condition)

a. Solve this model using the simplex method. b. Suppose, the objective function in this problem is profit ($) maximization [i.e. Maximize Z = 3x1 + 6x2]; solve this model using the simplex method.

3.

Consider the following linear programming problem. Maximize Subject to Z = 5x1 + 6x2 + 4x3 3x1 + 4x2 + 4x3 120 x1 + 2x2 + x3 50 x1 + 2x2 + 3x3 30 x1, x2, x3 0 The optimal simplex tableau is x1 Basis s3 x3 x1 zj cj - zj cj 0 4 5 5 0 0 1 5 0 x2 6 4 2 0.5 10.5 -4.5 x3 4 0 1 0 4 0 s1 0 -2 1 0 4 -4 s2 0 7 3 -2 2 -2 s3 0 1 0 0 0 0 Soln 80 30 20 220 1 2 3

a. Find the range of optimality for c1 and c2. b. Find the range of feasibility for b2. c. Suppose the right-hand side value of the first constraint is decreased from 120 to 110, what

will be the new optimal solution and its value? d. Suppose the right-hand side value of the second constraint is increased from 50 to 60, what will be the new solution and its value? e. What is the marginal increase (say in $) in the objective function if resource no 1 is increased by one unit? What is the amount of increase in the objective function if resource no 1 is increased by three units? f. The cost for a unit of resource no 1 is $2.50. Assume the objective function is profit ($), would you purchase the additional unit in resource no 1. Why?

Micro Computers Limited sells microcomputers to universities and colleges on the East Coast and ships them from three distribution warehouses. The firm is able to supply the following numbers of microcomputers to the universities by the beginning of the academic year;
4.

Distribution Warehouse 1. Richmond 2. Atlanta 3. Washington, D.C. Total

Supply (microcomputers) 420 610 340 1370

Four universities have ordered microcomputers that must be delivered and installed by the beginning of the academic year; Demand University A. Tech B. A and M C. State D. Central Total (microcomputers) 520 250 400 380 1550

The shipping and installation costs ($) per microcomputer from each distributor to each university are as follows; To From 1 2 3 A 22 15 28 B 17 35 21 C 30 20 16 D 18 25 14

a. Find the initial solution using;

i. NWCR ii. LCM, and iii. VAM b. Using the initial solution with NWCR method in (a) above, proceed to solve the problem for the optimal solution using MODI and Stepping Stone method. c. Suppose the demand in State is reduced from 400 units to 200 units, find the initial solution using NWCR. All other data remain the same. Proceed to solve the problem for the optimal solution using MODI and Stepping Stone method. 5. MS Consultant has received four consultancy works for the East Coast Development Corridor. The Executive Director is planning to assign the jobs to four of its senior consultants to lead each job so that the total completion time is minimized. The time frame (in weeks) for each of these consultants to complete each job is tabulated in the following table. Consultants A B C D Jobs 1 10 12 9 14 2 14 13 12 16 3 16 15 12 18 4 13 12 11 16

You have been given the task to assist the Executive Director to plan the job assignments. Find the optimal solution using the Hungarian method (Assignment method) so that the total completion time is minimized. a. State the optimal job assignments to each consultant. b. What is the total completion time for the consultancy works? c. Suppose, the firm has an additional consultant E, who can complete jobs 1, 2, 3, and 4 in 11, 15, 17 and 14 weeks respectively. What will be the optimal solution? The consultant not assigned to any of these jobs will be given a task to do Environmental Impact Assessment (EIA) study for other project. Which consultant will be assigned to this EIA study? d. Suppose, from the original problem, the firm has secured additional job 5, that can be completed by consultant A, B, C, and D in 15, 14, 13, and 17 weeks respectively. What will be the optimal solution? The job that has no consultant to do will be outsourced to another company. Which job will be outsourced?

6. Doug Casey is in charge of planning and coordinating next springs sales management training program for his company. Doug has listed the following activity information for this project. Immediate Activity Description Predecessor Optimistic Time (weeks) Most Probable A B C D E F G H I Plan topic Obtain speakers List meeting locations Select location Finalize speaker travel plans Make final check with speakers Prepare and mail brochures Take reservations Handle last-minute details A C B. D E B, D G F, H 1.5 2.0 1.0 1.5 0.5 1.0 3.0 3.0 1.5 2.0 2.5 2.0 2.0 1.0 2.0 3.5 4.0 2.0 2.5 6.0 3.0 2.5 1.5 3.0 7.0 5.0 2.5 Pessimistic

a. Draw a project network for this problem. b. Prepare an activity schedule. c. What are the critical activities (i.e. critical path) and what is the expected project completion time? d. If Doug wants a 0.99 probability of completing the project on time, how far ahead of the scheduled meeting date should he begin working on the project? e. If Doug wants 90% confidence to complete on time, how long will be the completion time? f. What is the probability for the project to be completed within 16 weeks? g. What is the likelihood (the probability) that the project is delayed beyond 17 weeks? h. Is it possible to delay the commencement of activities E & F? If yes, by how many weeks each.

ANSWERS 1. Answer a. LPP Formulation Let x1 = permanent operator x2 = temporary operator Min Z = 60x1 + 40x2 Subject to 24x1 + 16x2 600 x1 + x2 45 x1 + 2x2 60 x1, x2 0 minimize cost number of applications processed per day computer workstations available applications processed with errors per day non-negativity condition

b. Graphical solution Draw graph, grids 60 x1 and 60 x2

x2 60

50 Z 40

30

20 C 10

10

20

30

40

D 50

60

x1

Solution, A(25, 0) B(7.5, 26.25) C(30, 15) D(45, 0)

ZA = RM1500 ZB = RM1368.75 ZC = RM1975 ZD = RM2700

Therefore, Optimal Solution; ZBmin = RM1368.75 x1 = 7.5 (or 8 permanent operators) x2 = 26.25 (or 26 temporary operators) c. Solution, A(25, 0) B(7.5, 26.25) C(30, 15) D(45, 0)

ZA = RM1500 ZB = RM1500 ZC = RM2400 ZD = RM2700

Therefore, alternate optimal solution ZAmin = RM1500 x1 = 7.5 (8 permanent operators) x2 = 26.25 (26 temporary operators) ZBmin = 1500 x1 = 25 (25 or all permanent operators) x2 = 0 (0 or without temporary operators) 2. a.Minimization: Standard from: Minimize Z = 3x1 + 6x2 + 0s1 + 0s2 + 0s3 + 0s4 + Ma2 Subject to: 3x1 + 2x2 + s1 + 0s2 + 0s3 + 0s4 + 0a2= 18 x1 + x2 + 0s1 s2 + 0s3 + 0s4 + a2 = 5 x1 + 0s1 + 0s2 + s3 + 0s4 + 0a2 = 4 x2 + 0s1 + 0s2 +0s3 + s4 + 0a2 = 7 x1, x2, s1, s2, s3, s4, a2 0 Convert the minimization problem to maximization in order to proceed with the simplex method Let Z = Z(-1) Minimize Z = Maximize Z Thus: the minimization problem becomes maximization

Therefore the tableau form is Maximize Z = -3x1 - 6x2 - 0s1 - 0s2 - 0s3 - 0s4 - Ma2 Subject to: 3x1 + 2x2 + s1 + 0s2 + 0s3 + 0s4 + 0a2 = 18 x1 + x2 + 0s1 s2 + 0s3 + 0s4 + a2 = 5 x1 + 0s1 + 0s2 + s3 + 0s4 + 0a2 = 4 x2 + 0s1 + 0s2 +0s3 + s4 + 0a2 = 7 x1, x2, s1, s2, s3, s4, a2 0 Simplex method:
x1 Basis s1 a2 s3 s4 zj c j - zj s1 a2 x1 s4 zj c j - zj s1 x2 x1 s4 zj c j - zj 0 -6 -3 0 0 -M -3 0 cj 0 -M 0 0 -3 3 1 1 0 -M
-3+M

x2 -6 2 1 0 1 -M
-6+M

s1 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0

s2 0 0 -1 0 0 M -M 0 -1 0 0 M -M 2 -1 0 1 6 -6

s3 0 0 0 1 0 0 0 -3 -1 1 0 -3+M 3-M -1 -1 1 1 3 -3

s4 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 0 1 0 0

a3 -M 0 1 0 0 -M 0 0 1 0 0 0 -M 4 1 4 6 -18 6 1 4 7 -12-M 3 1 7 Soln 18 5 4 7 -5M Ratio 6 5 4 -

0 0 1 0 -3 0 0 0 1 0 -3 0

2 1 0 1 -M
-6+M

0 1 0 0 -6 0

Optimal solution (after three iterations: cj - zj 0): x1 = 4, x2 = 1, s1 = 4, s4 = 6

Zmax = -18 Thus, Zmin = Z(-1) = 18 x1 = 4, x2 = 1 b. Maximization: Standard from: Maximize Z = 3x1 + 6x2 + 0s1 + 0s2 + 0s3 + 0s4 - Ma2 Subject to: 3x1 + 2x2 + s1 + 0s2 + 0s3 + 0s4 + 0a2 = 18 x1 + x2 + 0s1 s2 + 0s3 + 0s4 + a2 = 5 x1 + 0s1 + 0s2 + s3 + 0s4 + 0a2 = 4 x2 + 0s1 + 0s2 +0s3 + s4 + 0a2 = 7 x1, x2, s1, s2, s3, s4, a2 0 Simplex method: Do it yourself (check your answer by using the simplex tool) Optimal solution (after five iterations): x1 = 4/3, x2 = 7 Zmax = 46 3. Sensitivity Analysis a. Range of optimality -4 c1, c1 6; -4 c1 6 c2 10.5
b.

Range of feasibility b2 b1 b2 . . bm a1j a2j . . amj

(original right hand side value, bm = 50)

bi

-10 b2 10 Range of feasibility for bi is 40 b2 60 (bm + bi) bi (bm + bi)

c.

New solution b1 b2 . . bm a1j a2j . . amj b*1 b*2 . . b*m

bi

Then find Znew

New optimal solution b1 = 110 120 = -10 x1 = 20 x3 = 20 s3 = 100 Znew = 180 d. New optimal solution b2 = 60 50 = 10 x1 = 40 x3 = 60 s3 = 150 Znew = 440

Marginal increase in the objective function resulting from an increase by one unit of resource no 1 is $4.00. [see dual price] Amount of increase in the objective function resulting from an increase of resource no 1 by three unit is $12.00
e.

Yes, because there is a net increase in the objective function of $1.50 per unit due to the increase in resource no 1 (i.e. Profit $4.00 less Cost $2.50)
f.

4.

Check whether the transportation problem is balanced

Supply = 1370; Demand = 1550 Supply Demand [difference 180 (i.e. 1550 - 1370)] The transportation problem is unbalanced; demand is 180 more than supply. Introduce Dummy Supply of 180 units with transportation cost of $0 per unit in each route. a. Initial (Hueristic) Solution NWCR
Universities Warehouses A 22 1 420 15 2 100 28 3 0 Dummy Demand 520 250 400 0 250 21 140 0 180 380 35 260 16 200 0 180 1550 14 340 20 25 610 B 17 C 30 D 18 420 Supply

Initial Solution: Total Cost: $29,730 Routes: 1-A=420; 2-A=100; 2-B=250; 2-C=260; 3-C=140; 3-D=200; Dummy-D=180

LCM
Universities Warehouses A 22 1 15 2 340 28 3 0 Dummy Demand 180 520 250 400 380 0 0 21 250 35 270 16 340 0 180 1550 14 340 B 17 130 20 C 30 40 25 610 D 18 420 Supply

Initial solution: Total Cost: $24,130 Routes: 1-B=250; 1-C=130; 1-D=40; 2-A=340; 2-C=270; 3-D=340; Dummy-A=180

VAM
Universities Warehouses A 22 1 15 2 520 28 3 0 Dummy Demand Column penalty 520 15,7,-,-, -,-,180 250 17,4,4,4, -,-,400 16,4,4,14, -,-,380 14,4,4,4, 4,4,7 0 21 310 0 70 35 90 16 30 0 180 1550 14 340 20 B 17 C 30 350 25 610 D 18 420 Supply row penalty 1,1,1,1, 1,1,1 5,5,5,-, -,-,-, 2,2,2,2, 2,7,0,-,-,-, -,-,-

Initial Solution: Total Cost = $22,470 Routes:1-B=70; 1-D=350; 2-A=520; 2-C=90; 3-C=310; 3-D=30; Dummy-B=180

b. MODI and Stepping Stone Method:

Check whether m + n 1 = number of basic cells m+n1=4+41=7 number of basic cells = 7 Thus; m + n 1 = number of basic cells Therefore we can proceed with MODI and Stepping Stone Method First iteration; find auxilary indices & net evaluation indices (improvement indices) Check net evaluation indices, whether it is optimal. In this case, it is not, therefore proceed with close path for MODI & Stepping Stone
Universities Warehouses A 22 1 420 15 2 100 + 28 3 (17) 0 Dummy Demand vj (3) 520 22 (-17) 250 42 (-10) 0 (-2) 400 27 250 (-25) B 17 + 35 21 140 0 180 380 25 260 16 200 0 180 1550 -25 (3) 20 (7) 14 340 -11 C 30 (-7) 25 610 -7 D 18 Supply 420 ui 0

Second iteration; find auxilary indices & net evaluation indices (improvement indices)
Universities Warehouses A 22 1 170 15 2 350 + 28 3 (17) 0 Dummy Demand vj (3) 520 22 (8) 250 17 (15) 0 (-2) 400 27 (25) 21 140 250 35 260 B 17 (3) 20 16 + 0 180 380 25 200 (7) 14 340 0 180 1550 -25 -11 C 30 (-7) D 18 + 25 610 -7 Supply 420 ui 0

Third iteration; find auxilary indices & net evaluation indices (improvement indices)
Universities Warehouses A 22 1 (7) 15 2 520 28 3 (17) 0 Dummy Demand vj (3) 520 15 (1) 250 17 (8) 0 (-2) (18) 21 310 250 35 90 16 0 + 400 20 180 30 B 17 (10) 20 (7) 14 340 + 0 380 18 180 1550 -18 -4 C 30 170 25 610 0 D 18 Supply 420 ui 0

Fourth iteration; find auxilary indices & net evaluation indices (improvement indices)
Universities Warehouses A 22 1 (7) 15 2 520 28 3 (17) 0 Dummy Demand vj (5) 520 15 (3) 250 17 (8) 0 180 400 20 (18) 21 130 0 (2) 380 18 250 35 90 16 210 0 180 1550 -20 B 17 (10) 20 (7) 14 340 -4 C 30 170 25 610 0 D 18 Supply 420 ui 0

On fouth iteration, the net evaluation indices (or improvement indices) 0 Thus, on fifth iteration, reach optimal solution; The optimal solution is; Total Cost = $21,930 Optimal routes: 1-B=250; 1-D=170; 2-A=520; 2-C=90; 3-C=130; 3-D=210; DummyC=180

c.

If demand in State is reduced to 200, total demand = 1350, while total supply =

1370. Total supply Total demand; i.e. unbalanced transportation problem: total supply is less than total demand by 20 (1370 1350 = 20). Therefore introduce Dummy Demand of 20 units with transportation cost of $0 per unit in each route. NWCR
Universities Warehouses A 22 1 420 15 2 100 28 3 Demand 520 250 200 250 21 35 200 16 320 380 20 60 14 20 20 0 340 1370 25 0 610 B 17 C 30 D 18 Dummy 0 420 Supply

Initial Solution: Total Cost: $29,730 Routes: 1-A=420; 2-A=100; 2-B=250; 2-C=200; 2-D=60; 3-D=320; 3-Dummy=20 Proceed with MODI & Stepping Stone Method. m + n 1 = 7; number of basic cells = 7 Therefore: m + n 1 = number of basic cells So you can proceed with MODI and Stepping Stone Method Do it yourself, using exactly the same procedure: find the values of auxiliary indices and net evaluation indices (improvement indices) and then check for optimality at every iteration until optimal solution is obtained (net evaluation indices 0). Then state your solution: Total Cost and the Optimal Routes.

5. a. To determine job assignments: Check whether assignment problem is balanced. Number of consultants = 4; number of jobs = 4; thus, assignment problem is balanced [number of consultants = number of jobs]. Therefore, we can proceed with the solution using Hungarian method without adding dummy consultant or dummy job. [if number of consultant is more, introduce dummy job with completion time 0 for each consultant; if number of job is more, introduce dummy consultant with completion time 0 for each job] The assignment tableau Jobs 1 10 12 9 14 2 14 13 12 16 3 16 15 12 18 4 13 12 11 16

Consultants A B C D

Row operation (compute opportunity costs row-wise) Jobs Consultants 1 2 A B C D 0 0 0 0 4 1 3 2

3 6 3 3 4

4 3 0 2 2

Column operation (compute opportunity costs column-wise) Jobs Consultants 1 2 3 A B C D 0 0 0 0 3 0 2 1 3 0 0 1

4 3 0 2 2

First iteration: Hungarian Method [draw minimum number of lines covering zeros] Jobs Consultants 1 2 3 4 A B C D 0 0 0 0 3 0 2 1 3 0 0 1 3 0 2 2

number of lines number of rows or columns; Therefore not optimal Second iteration: Re-computation of opportunity cost [adjust] Jobs Consultants 1 2 3 A B C D 0 1 0 0 2 0 1 0 3 1 0 1

4 2 0 1 1

Second iteration: Hungarian method [draw lines] Consultants A B C D Jobs 1 0 1 0 0 2 2 0 1 0 3 3 1 0 1 4 2 0 1 1

no of lines = no of rows or columns; Therefore optimal solution is reached Assign consultant to jobs; i.e. Optimal assignments Jobs Consultants 1 2 A B C D 0 0 0 0 0 0 0

The assignments: Consultant A to Job number 1 with completion time of 10 weeks Consultant B to Job number 4 with completion time of 12 weeks Consultant C to Job number 3 with completion time of 12 weeks Consultant D to Job number 2 with completion time of 16 weeks b. Total completion time is 10 + 12 + 12 + 16 = 50 weeks
c. Consultants are more than Jobs (unbalanced assignment problem)

Do it yourself. Introduce dummy job with completion time of 0 for each consultant. Alternative optimal assignment: A-2; B-4; C-3; D-Dummy; E-1: or A-1; B-4; C-3; DDummy; E-2: or A-4; B-2; C-3; D-Dummy; A-1. Completion time: 49 weeks Consultant D will be assigned to conduct EIA study.
d. Jobs are more than Consultants (unbalanced assignment problem)

Do it yourself. Introduce dummy consultant with completion time of 0 for each job. Optimal assignment: A-1; B-4; C-3; D-2; Dummy-5. Completion time: 50 weeks Project number 5 will be outsourced.

6. a. Network diagram

Start

Finish

e. Activity schedules Activity A B C D E F G H I Expected Time 2 3 2 2 1 2 4 4 2 Variance 0.03 0.44 0.11 0.03 0.03 0.11 0.44 0.11 0.03

Critical activities Earliest Activity A B C D E F G H I Start 0 2 0 2 5 6 5 9 13 Latest Start 0 2 1 3 10 11 5 9 13 Earliest Finish 2 5 2 4 6 8 9 13 15 Latest Finish 2 5 3 5 11 13 9 13 15 Slack 0 0 1 1 5 5 0 0 0 Yes Yes Yes Critical Activity Yes Yes

c. Critical Path: Start A B G H I Finish Expected completion time, E(T) = 2 + 3 + 4 + 4 + 2 = 15 weeks d. Variance on critical path 2 = 0.03 + 0.44 + 0.44 + 0.11 + 0.03 = 1.05 From Probability Distribution Table, we find 0.99 probability occurs at z = +2.33. Thus
z = T - E (T) = T - 15 = 2.33 1.05 or T = 15 + 2.33 1.05 = 17.4 weeks

e. f.

T = 15 + 1.28 1.05 = 16.3 weeks P(complete 16 weeks) = (0.5 + 0.3353) = 0.8353 or 83.53%

g. Z = 1.9518, Therefore P(complete 17 weeks) = (0.5 0.4744) = 0.0256 or 2.56% Yes, commencement of both activities can be delayed by 5 weeks each (slack is 5 for each activity)
h.