University of Engineering and Technology, Peshawar

Department of Industrial Engineering

Fall 2020, IE-356 Operations Research
Assignment No. 3
Complex Engineering Problem
Due Date: 24th Feb, 2021 Total Marks: 100

Submitted by: Ziaullah Jan, Registration # 18pwind0541, Section: A

Question:

An oil refinery produces four types of raw gasoline: Alkylate, Catalytic-Cracked, straight-run, and Isopentane.
Two important characteristics of each gasoline are its performance number PN (including anti knock properties)
and its vapor pressure RVP (indicating volatility). These two characteristics, together with the production levels
in barrels per day, are as follows:
Table 1 RAW Gasolines Characteristics and Production Levels

PN RVP Barrels Produced

Alkylate 107 5 3,814

Catalytic-cracked 93 8 2,666

Straight-run 87 4 4,016

Isopentane 105 21 1,300

These gasoline’s can be sold either raw, at $4.83 per barrel, or blended into aviation gasolines (Avgas A and/or
Avgas B). Quality standards impose certain requirements on the aviation gasolines: these requirements, together
with the selling prices, are as follows
Table 2 Aviation Gasolines Characteristics and Prices

PN RVP Price Per Barrel

Avgas A At least 100 At most 7 $6.45

Avgas B At least 91 At most 7 $5.91

The PN and RVP of each mixture are simply weighted averages of the PN and RVPs of its constituents.

The refinery has three units having daily capacities of 6, 5, and 8 million gallons, respectively, supply to three
distribution areas with daily demands of 4, 8, and 7 million gallons, respectively. Gasoline is transported to the
three distribution areas through a network of pipelines. The transportation cost is 10 cents per 1000 gallons per
pipeline mile. Table 3 gives the mileage between the refineries and the distribution areas.
Table 3 Mileage between refineries and the distribution areas.

Distribution Area 1 Distribution Area 2 Distribution Area 2

 Refinery 1 180 180 2000

Refinery 2 300 800 900

Refinery 3 220 200 120

a. Formulate an LP Model only for the transportation aspect of the problem. (15 Marks) (C4, Problem
Analysis)
b. Use any method/software to identify the optimum shipping schedule. (15 Marks) (C3, Engineering
Knowledge)
c. Evaluate and compare the answers with actual problem if the supply to Area 3 is increased by 5 million
gallons. (15 Marks) (C6, Problem Analysis)
d. Transform the problem into a network model (15 Marks) (C4, Problem Analysis) (Note: You can use
any assumption for this part provided all assumptions are clearly mentioned.)
e. If there is 1-1 assignment for Table 3. Use Hungarian Method to solve the problem (10 Marks) (C3,
Engineering Knowledge).
f. Refinery 1 serves 3 tankers at a time and provides parking space to 8 tankers. If the space is full the
tankers are not allowed to enter the facility. The tankers arrivals occur according to Poisson distribution
with a mean of four per hour. The time get the tanker filled is exponential with mean 15 minutes. Assess
the probability that a tanker will not be allowed to enter the facility (15 Marks) (C6, Problem Analysis).
g. Considering the LP Model in part (a), evaluate and interpret the dual prices of all the resources under
consideration. (15 Marks) (C6, Problem Analysis)

Solution
a. Formulate an LP Model only for the transportation aspect of the problem.
 b. Use any method/software to identify the optimum shipping schedule.

I have used TORA to solve this transportation model. North west corner method is used to find the optimum
schedule for this model.

Data entry in TORA

Figure 1 Data entry in TORA

Solution

Figure 2 Optimal Iteration

 Figure 2 shows that in the 2nd iteration we get the optimal solution.

Figure 3 Optimum Schedule

Objective value of this model

Objective value for this model is $41.20

 c. Evaluate and compare the answers with actual problem if the supply to Area 3 is increased by 5
million gallons.

We will use the same TORA software and increase the supply to area 3 by 5 million gallons.

Data entry

Figure 4 Data entry in TORA

Supply to area 3 has now become 13 million.

Solution

We will use the same method as we did in the last part.

We get the optimal solution in 2nd iteration.

Now let’s find the schedule then we will evaluate it with the last part.
 Figure 5 Optimum schedule

From this schedule we get the Objective value = 61.20

Evaluation & comparison

We can see from both the solutions that changing the demand of 1 distribution area had an effect on objective
value. Increasing the demand has increased the objective value which is actually the total cost.

Secondly the schedule has also changed. Pattern of supply of the demands have changed. In first part in 2 nd
iteration refinery 1 supplies the demand to distribution 2, while in the 2 nd part refinery 1 supplies to distribution
1 and 2.

So we can conclude from above solutions that changing the supply have an increase on total cost as well as
optimum schedule.
d. Transform the problem into a network model

4
Refinery Distn
1 area 1
6

1
Distn
Refinery
area 2
2

1
Distn
Refinery
area 3
3
7

Figure 6 Network model

 e. If there is 1-1 assignment for Table 3. Use Hungarian Method to solve the problem.

Solution
f. Refinery 1 serves 3 tankers at a time and provides parking space to 8 tankers. If the space is full the
tankers are not allowed to enter the facility. The tankers arrivals occur according to Poisson distribution
with a mean of four per hour. The time get the tanker filled is exponential with mean 15 minutes. Assess
the probability that a tanker will not be allowed to enter the facility.
 g. Considering the LP Model in part (a), evaluate and interpret the dual prices of all the resources under
consideration.

Now we will solve the Linear programme in part a by simplex method in TORA to find the dual prices.

Steps

Step 1 Enter the LP model in TORA

Step 2 This is the first iteration where we make artificial variables in obj. function zero
Step 3 Identify pivot column, row and element in iteration 1. Then move towards iteration 2 where we make
pivot element 1, elements above and below it zero.

Step 4 Similarly in iteration 3 identify above mentioned pivot element row and column in iteration 2 and then
perform row operations on it.

Step 5 This is 4th iteration

Step 6 Iteration 5th

Step 7 This is 6th and final iteration

Dual price

Now we will evaluate and interpret dual prices for each resource from the 6 th iteration tableau.

Resource Dual Price

Constraint 1 1

Constraint 2 7.20

Constraint 3 1.20

Constraint 4 -7.20

Constraint 5 0.80

Constraint 6 0.00