VIETNAM NATIONAL UNIVERSITY – HO CHI MINH CITY
INTERNATIONAL UNIVERSITY
SCHOOL OF INDUSTRIAL ENGINEERING & MANAGEMENT
SEMESTER PROJECT
Course: Deterministic Models in Operations Research
Instructor: Dr. Nguyen Van Hop
Group 11:
Le Phuong Ngoc Anh IEIEIU18004
Tran Anh Thu IEIEIU18160
Pham Ngoc Thu Uyen IEIEIU18114
Dinh Phu Vinh IEIEIU18118
June 2020
TABLE OF CONTENTS
I. PROBLEM IDENTIFICATION .......................................................................................... 2
1. Problem ................................................................................................................................ 2
2. Data summary ...................................................................................................................... 2
3. Objective .............................................................................................................................. 3
4. Scope of the project ............................................................................................................. 4
5. Limitation ............................................................................................................................. 4
II. MODEL DEVELOPMENT .............................................................................................. 4
III. PROBLEM SOLVING ...................................................................................................... 5
1. Lingo Model......................................................................................................................... 5
2. Solution by LINGO.............................................................................................................. 6
3. Sensitivity analysis............................................................................................................... 7
IV. APPLICATIONS ............................................................................................................... 9
1. Case study ............................................................................................................................ 9
2. Model ................................................................................................................................... 9
3. Solution ................................................................................................................................ 9
V. CONCLUSION ................................................................................................................ 10
1
I. PROBLEM IDENTIFICATION
1. Problem
The Weigelt Corporation has three branch plants with excess production capacity.
Fortunately, the corporation has a new product ready to begin production, and all three
plants have this capability, so some of the excess capacity can be used in this way.
This product can be made in three sizes -- large, medium, and small -- that yield a net unit
profit of $420, $360, and $300, respectively. Plants 1, 2, and 3 have the excess capacity
to produce 750, 900, and 450 units per day of this product, respectively, regardless of the
size or combination of sizes involved. The amount of available in-process storage space
also imposes a limitation on the production rates of the new product. Plants 1, 2, and 3
have 13,000, 12,000, and 5,000 square feet, respectively, of in-process storage space
available for a day’s production of this product. Each unit of the large, medium, and
small sizes produced per day requires 20, 15, and 12 square feet, respectively. Sales
forecasts indicate that if available, 900, 1,200, and 750 units of the large, medium, and
small sizes, respectively, would be sold per day.
At each plant, some employees will need to be laid off unless most of the plant’s
excess production capacity can be used to produce the new product. To avoid layoffs if
possible, management has decided that the plants should use the same percentage of their
excess capacity to produce the new product. Management wishes to know how much of
each of the sizes should be produced by each of the plants to maximize profit.
Formulate and solve a linear programming model for this problem.
2. Data summary
Branch Excess capacity to In-process storage space available for a day’s
plants produce (units/day) production (square feet)
Plant 1 750 13,000
Plant 2 900 12,000
Plant 3 450 5,000
*Plants should use the same percentage of their excess capacity to produce the new
product
Size Net unit profit Space size to Sales forecasts if available
produce (square (units)
feet)
2
Large $420 20 900
Medium $360 15 1,200
Small $300 12 750
3. Objective
At first, our group will find optimal solution to the problem. Following the
overall optimization paradigm, when applying research, a decision-maker will select the
key decision variables that influence the full quality of choices. These are expressed
by the objective function: maximized profit. Additionally to the function, a collection of
physical, technical, economic, environmental, legal, societal, etc. constraints considered.
By adjusting the values of all decision variables, a decent (feasible) or absolute
best (optimal) solution is chosen. Feasibility and optimality can only be defined within
the context of the given problem formulation.
After solving the values, we will examine some figures in order to help company to
determine sensitivity analysis of this model through some factors:
- Reduced costs: Are the values that we can reduce in coefficients of objective
function so that their variable become basic variables. The reduced cost portion of
the LINGO output gives us information about how difference the objective
function coefficient for a nonbasic variable will change the LP’s optimal solution.
- Shadow price: As we will see in some cases, it is often important for managers to
determine how constraint’s right-hand side changes the LP’s optimal z-value.
With this in mind, we define the shadow price for the ith constraint of an LP to be
the amount by which the optimal z-value is improved-increased in a max problem
if the right-hand-side of the ith constraint is increased by 1.
- Range of coefficient in objective funtion: We can determine (at least in a two-
variable problem) the range of values for an objective function coefficient for
which the current basis remains. For each objective function coefficient, this range
is given in the objective coefficients ranges portion of the LINGO output. The
allowable increase (AI) section indicates the amount by which an objective
function coefficient can be increased with the current basis remaining optimal.
Similarly, the allowable decrease (AD) section indicates the amount by which an
objective function coefficient can be decreased with the current remaining
optimal.
3
- Range of right-hand-side: We can determine (at least for a two-variable problem)
the range of values for a right-hand side within which the current basis remains
optimal. This information is given in the right-hand-side ranges section of the
LINGO output. Even if a change in the right-hand side of a constraint leaves the
current basis optimal, the LINGO output does not provide sufficient information to
determine the new values of the decision variables. However, the LINGO output
does allow us to determine the LP’s new optimal z-value.
4. Scope of the project
In this problem, we will examine max profit through capacity, net profit, storage
space and sales forecasts constraints in LINGO 17.0. Subsequently, we use Lingo to
analyze the sensitivity of the problem. Sensitivity analysis is important for some reasons.
In many applications, the values of an LP’s parameters may change. For example, the
prices at which soldiers and trains are sold or the availability of carpentry and finishing
hours may change. If a parameter changes, by using sensitivity analysis, it may be
unnecessary to solve the problem again. For example, if the profit contribution of a
soldier increased to $3, we would not have to resolve the problem, because the current
solution remains optimal. A knowledge of sensitivity analysis often enables the manager
to determine from the original solution how changes in an LP’s parameters change its
optimal solution.
5. Limitation
When considering bigger scale, the output may not be optimal. So, if we apply this
case into real life problems, for example when manager using this method to production
and supplyment, some statics and data must be added like capacity, cost, delay,
labor…and may be rebuilt those models again. By doing this, the output will have better
solution as well as the accuracy and efficiency of the task although it may take more
time. Additionally, those figures in the given problem are integers so that we can not
determine shadow price for this model.
II. MODEL DEVELOPMENT
Let Xij be the number of units produced per day at plant i (with i = 1, 2, 3) and with
size j product (j = L, M, S = large, medium, small).
Let Z be the total net profit per day, the mathematic model for this problem is shown
below:
The objective function is:
4
Max Z = 420(X1L + X2L + X3L) + 360(X1M + X2M + X3M) + 300(X1S + X2S + X3S)
subject to following constraints:
Capacity constraints:
X1L + X1M + X1S 750 (1)
X2L + X2M + X2S 900 (2)
X3L + X3M + X3S 450 (3)
Storage constraints:
20X1L + 15X1M + 12X1S 13000 (4)
20X2L + 15X2M + 12X2S 12000 (5)
20X3L + 15X3M + 12X3S 5000 (6)
Sales forecasts constraints:
X1L + X2L + X3L 900 (7)
X1M + X2M + X3M 1200 (8)
X1S + X2S + X3S 750 (9)
Equal capacity percentage constraints:
900(X1L + X1M + X1S) – 750(X2L + X2M + X2S) = 0
450(X2L + X2M + X2S) – 900(X3L + X3M + X3S) = 0
Non-negative constraints:
All Xij 0
Integral constraints:
All Xij are integer.
III. PROBLEM SOLVING
We use Lingo to solve the problem and analyze the problem sensitivity.
1. Lingo Model
5
2. Solution by LINGO
- Model class: Pure integer linear programming
- Solver type: Branch-and-Bound
6
Maximum profit: $694,680, produce 530 units of large size in plant 1,
160 units of medium size in plant 1,
140 units of small size in plant 1,
688 units of medium size in plant 2,
404 units of small size in plant 3,
10 units of medium size in plant 3 per day.
3. Sensitivity analysis
Using Lingo and trial-and-error method, we obtain:
- Reduced cost:
Reduced
No. Variable
cost
1 X1L 0
2 X2L 41
3 X3L 1
4 X1M 0
5 X2M 0
6 X3M 0
7 X1S 25
8 X2S 0
9 X3S 0
- Ranges in which the basis is unchanged:
Objective Coefficient Ranges:
7
Current Allowable
No. Variable
Coefficient Change
1 X1L 420 -90
2 X2L 420 41
3 X3L 420 1
4 X1M 360 -16
5 X2M 360 -178
6 X3M 360 -15
7 X1S 300 25
8 X2S 300 -84
9 X3S 300 -300
Righthand Side Ranges:
No. of Current Allowable Allowable
constraint RHS Increase Decrease
1 750 infinity 60
2 900 infinity 72
3 450 infinity 34
4 13000 1 N/A
5 12000 2 N/A
6 5000 1 N/A
7 900 infinity 370
8 1200 infinity 342
9 750 infinity 206
Range of RHS of constraints to keep basic variables remain optimal:
(1): 690 RHS <
(2): 828 RHS <
(3): 416 RHS <
(4): 13000 RHS 13001
(5): 12000 RHS 12002
(6): 5000 RHS 5001
(7): 530 RHS <
(8): 858 RHS <
(9): 530 RHS <
8
IV. APPLICATIONS
1. Case study
The West Hartford Senior Center is trying to establish a transportation system of
small and large vans. It can spend no more than $100,000 for both sizes of vehicles
and no more than $500 per month for maintenance. The WHSC can purchase a small
van, which carries up to 7 passengers for $10,000 and maintain it for $100 per month.
The large vans, which carry up to 15 passengers, cost $20,000 each and can be
maintained for $75 per month. How many of each type of van should they purchase if
they want to maximize the number of passengers?
2. Model
Let
So, we have the model:
3. Solution
Maximum: 75 passengers, purchase only 5 large vans.
9
V. CONCLUSION
In conclusion, we have the solution that is The Weigelt Corporation should produce
530 of large size, 160 of medium size, 140 of small size on plant 1; 688 of medium size;
10 of medium size, 404 of small size on plant 3, with the maximum profit P = $694,680.
We also examine the sensitivity of this problem. Sensitivity analysis provides a method to
evaluate the amount of risk involved in a project. Sensitivity analysis also involves
calculating the impact of variations on different assessable components of a project,
helping manager identify potential plans. Manager may also use sensitivity analysis to
identify which components of a plan when changing slightly, will impact the outcome of
a project most.
Thus, this is a practical problem applied mathematical methods of principle and
objectives of Operations Research to maximize its profit. Also, it gives students a chance
to apply our knowledge and understanding easily to make better decisions for the future
career.
10
