Selfservice gas station
 Project name: Selfservice gas station
 Class: 4IT496 Simulation of Systems (WS 2013/2014)
 Author: Bohuslav Dvorský
 Model type: Descreteevent simulation
 Software used: SimProcess
Contents
Introduction
In Western countries becoming increasingly popular selfservice gas stations  a station where is no staff, but only machine that after the payment distribute fuel or other goods like mix washed, distilled water etc… With my friend we were thinking about a similar concept in the Czech conditions and realistically began to explore the possibility of implementation of the business plan. We contacted fuel distributor, manufacturer that would provide us station infrastructure (fuel tank, tubes, software, fuel dispenser…), directorate for roads and highways, a construction company that would build asphalt road connector, many banks for best offer of cashless payment and business loan and more other institutes. So we got estimate price of everything we need for construction of selfservice gas station including payment for permissions from different authorities. Thanks acquaintances of my friend, we also have actual data from the gas station standing on the same path where we would like to build our station. It is the Benzina gas station and we have data like average number of cars on the station, average liters of fuel tanked, ration of diesel, petrol and LPG.
Problem definition
Can be this project profitable? As we discovered, more than 50% of people use credit card to pay the fuel. Unfortunately banks for companies like ours charge a fee for their services in the amount of 1.48% of the total. This finding is very frustrating simply because the margin for fuel out of a large network of filling stations is about 2 CZK per liter. This finding is very frustrating simply because the margin for fuel out of a large network of filling stations is about 2 CZK per liter. So most of the revenue is likely to be cash payments or loyalty card payments. Thanks to the selfservice station, we can eliminate the cost of human factor, main building or restrooms. But when considering starting a new business we muse take a loan in the amount of almost 100% of the capital. We must also include the cost of bank loan, electricity, internet and maintenance of the station.
I believe that in certain circumstances the project is profitable. My job is to find the conditions under which selfservice station made a profit and paid itself.
What do we know?
As I stated in the introduction, we have obtained all the information to which we could get and made contact with the fuel supplier.
 The lowest possible price for the station: [ CZK ] 3000000
 Price for another dispenser / connection to the system: [ CZK ] 250000
 Price for a split tank with diesel / petrol: [ CZK ] 800000
 Price for road connector in the chosen location: [ CZK ] 1500000
 Price for the parcel: [ CZK ] 500000
 Monthly fixed costs  electricity, internet, maintenance: [ CZK ] 35000
 The cost of a dispenser  asphalt road: [ CZK ] 200000
 Capacity of the tank: [ l ] 15000
 basic type of station (lowest possible price) include: Fuel tank, fuel dispenser (diesel/petrol), infrastructure, asphalt road, software...everything needed for running the station
 split tank: fuel tank is divided into two sections  7500 litres of diesel and petrol
 Average number of customers: 502
 Average litres of fuel tanked: 35
 Diesel/Petrol tanked ratio: 60% / 40%
 Payments ratio  credit card/cash/loyalty card: 67% / 14% / 19%
 Fee for credit card payment: 1.48% of total amount
We obtained this data from the employee of Benzina gas station in Šterbnerk. This gas station is situated in the same locality as our fictive gas station so we can label this data as relevant for us.
LPG can be refueled only by trained personnel which for this type of station is unacceptable.
Fuel supplier
We communicated with the nearest fuel supplier that meets our conditions. Because our station is located within a distance of 35 km from the headquarters of the supplier, the price per delivery is included in the price of fuel. Our supplier is company Albeva Morava s.r.o.. Our conditions were: price for delivery included in price of fuel and just in time delivery with continuous replenishment of fuel. The idea is that the station software will detect low fuel level and alert our supplier who then begins the process of delivery of the fuel. We were asured that our station software can send email or it can be connected to ERPlike system of various suppliers. Email variant was enough for our suplier and further terms would be specified in the contract. So we found another question...what level of fuel should warn our supplier due to the variable cost for the supply of different quantities of fuel?
Fuel delivery (price for diesel): [ l ]
10001 15000 : 34,48 [ CZK ]
15001  20000 : 34,24 [ CZK ]
20001  X : 33,81 [ CZK ]
Fuel delivery (price for petrol): [ l ]
10001 15000 : 34,03 [ CZK ]
15001  20000 : 33,91 [ CZK ]
20001  X : 33,79 [ CZK ]
Location
A parcel of size aproximately 1600 square metres. Price for metre square is 300 CZK. This parcel is situated on the main way from Šternberk to Opava and also to Poland. So there are plenty of cars heading to Poland or to Opava.
Method
Lets start with simple math. We know that Benzina has aproximately 502 customers per day and they tank 35 litres per average. That is 17570 litres per day and divided with diesel/petrol ratio we get 7028 litres of petrol and 10542 litres of diesel. With fuel divided to petrol and diesel we can now calculate the costs of fuel.
We have three variants due to volume of petrol delivery:
Variant A : 10001  15000 litres
Diesel cost: 363488,16 [ CZK ]
Petrol cost: 239162,84 [ CZK ]
Variant B : 150001  20000 litres
Diesel cost: 360958,08 [ CZK ]
Petrol cost: 238319,48 [ CZK ]
Variant C : 20001  X litres
Diesel cost: 356425,02 [ CZK ]
Petrol cost: 237476,12 [ CZK ]
After consulting with a financial advisor from Fincentrum.cz I was told that it is possible to get a business loan with an interest rate of around 5.5% for 5 years, 6% loan for 10 years and 6.5% for 15 years. So lets sum what we know about our fixed costs and cost for building the station. We consider the smallest possible size of the company.
 3 000 000 CZK  complete system
 800 000 CZK  aditional fuel tank (if capacity is 15000 litres and Benzina sell 17000 litres per day we sure need one more)
 500 000 CZK  parcel
 1 500 000 CZK road connector
 700 000 CZK  savings
Total = 6 500 000 CZK
Now we can calculate monthly installment.
5,5% interest rate  5y loan
= 124 157,55 CZK
6% interest rate  10y loan
= 72 163,33 CZK
6,5% interest rate  15y loan
= 56 621,98 CZK
And with monthly installments we can define total fixed costs:
Variant 5.5
 124 157,55 + 35 000 = 159 157,55 CZK
Varian 6
 72 163,33 + 35 000 = 107 163,33 CZK
Variant 6,5
 56 621,98 + 35 000 = 91 621,98 CZK
We have estabilished that for profit we must make more money than our fixed costs
How much does Benzina earn?
 price for their diesel is 36,90 and price for petrol is 36,50
 we dont know anything about their fee for credit card payment  it sure be lower than ours
 we dont know anything about their price for fuel from supplier  it sure be lower than ours but now we assume our prices
But with numbers above we can say that their gross margin for diesel is 32574,78 CZK and gross margin for petrol is 19045,88 CZK.
Total gross margin 51620,66 CZK per day.
Can we make our net margin sufficiently high to beat our fixed cost?
The first thing we have say is that we can not have such high prices as Benzina. People will not believe the new technology and will rely on traditional methods. People will not believe the new technology and will rely on traditional methods. The only attraction for them will therefore price and quality of service.
Lets assume that we set prices to 36,00 for diesel and 35,60 for petrol and values from our example. It is gonna to be profitable?
 net margin for diesel: gross margin  bank fees = 19323,73901 CZK
 net margin for petrol: gross margin  bank fees = 10239,72853 CZK
 total margin per day = 29563,46754 CZK
It seems like this project is profitable. But this is only simple math. The is no guarantee that this station would have so many customers even with lower prices. Also we have only one fuel dispenser so there will we definitely queue and angry drivers will leave after few minutes and never come back. So that is what I am gonna simulate.
Method of this simulation is to use SimProcess to create virtual gas station and simulate diferent conditions to achieve real results
So there will different number of customers every time, different number of fuel dispensers and different price.
Model
Stages
 Arrival
 Fueling
 Exit
Entities
 Customer
Resources
 fuel dispenser
 diesel
 petrol
Arrival stage
Thanks to data from Benzina, we know when peaks occur. So we can estabilish a generator that will provide us customers according to real data. There are 6 different time perionds with different "breed rate".
Fueling stage
This stage compose of Branch that divide customers to those who tank diesel and those who tank petrol. There is Delay that is simulating the fueling proces. We measured that ordinary person is done with fueling and payment in Nor(140,20). This Delay also provide us the possibily to simulate angry customers who are leaving after 10 minutes of waiting. There is also Get Resource process that simulate fueling of petrol or diesel. Everytime when customer needs to refuel he needs the fuel dispenser. There is limited number of fuel dispensers and their release is controled by Release Resource proces.
Exit stage
Nothing special.
Scenario 1
= GREAT OPTIMIST
 average 600 customers per day
 huge peak 05:00  11:00 + 15:00  17:00
 2 fuel tanks
We surely need to keep customer satisfaction on the highest level possible. That mean no queues and low prices.
Lets try it with one fuel dispenser.
As you can see in one day visited station 601 customers. Unfortunately, it queues up to 14 customers and there were moments when the customer waited longer than 10 minutes. 

There were also a great number of angry customers that leave. 94 that wanted diesel and 84 that wanted petrol....result? = NOT OPTIMAL
Now lets try 2 fuel dispensers. ( + 400 000 CZK for fuel dispenser and asphalt road)
Generaly the result is not bad. We managed to lower the number of angry customers to total 6. But it is still not perfect. Customers are waiting too long for dispenser and their "visit rate" could decrease in time.
Lets try 3 fuel dispensers. ( + 800 000 CZK for fuel dispenser and asphalt road)
Optimal result. No angry customers. Customers are not waiting for a long time. Lets do math and se how is this scenario profitable:
a) miracle  diesel 36,00 CZK and petrol 35,60 CZK
Litres of fuel sold:
 12 516 litres of diesel
 8 344 litres of petrol
Net margin (diesel + petrol): 35099,25628 CZK
Monthly income: 1052977,688 CZK
= PROFITABLE FOR EVERY LOAN VARIANT
b) attractive price  diesel 35,10 and petrol 34,70
Litres of fuel sold:
 12 516 litres of diesel
 8 344 litres of petrol
Net margin (diesel + petrol): 16511,41927 CZK
Monthly income: 495342,578 CZK
= PROFITABLE FOR EVERY LOAN VARIANT
Scenario 2
= BENZINA LIKE
 average 500 customers per day
 huge peak 05:00  11:00 + 15:00  17:00
 2 fuel tanks
Lets assume that we can due to lower prices estabilish the same "visit ratio" as Benzina. To do this we need to keep the customer satisfaction on high level. That mean no queues, short waiting time and low prices.
We will start with 1 fuel dispenser.
Barely optimal. One fuel dispenser is simply not enough.
Lets try 2 fuel dispensers. ( + 400 000 CZK for fuel dispenser and asphalt road)
As we can see, we are getting closer to optimal state. There are 504 customers in the system and 8 customers max waiting for fuel dispenser. No one leaves due to angrer of waiting. That is practicaly real situation in rush hour. But still customer who is waiting for 9,2 minutes have to be pretty frustrated. For the sake of future profit we need to be more open and flexible to the needs of our clients.
Now we add one more fuel dispenser. Total number = 3. ( + 800 000 CZK for fuel dispenser and asphalt road)
In this case we achieved optimal state. There are max 3 customers waiting in the queue for max 3,4 minutes. No one leaves with anger and 3 minutes of waiting is a reasonable amount of time spent on wainting for cheap petrol or diesel. Lets find out if the scenario is profitable
a) high prices  diesel 36,00 CZK and petrol 35,60 CZK
Litres of fuel sold:
 10521 litres of diesel
 7014 litres of petrol
Net margin (diesel + petrol): 29504,57617 CZK
Monthly income: 885137,2851 CZK
= PROFITABLE FOR EVERY LOAN VARIANT
b) attractive price  diesel 35,10 and petrol 34,70
Litres of fuel sold:
 10521 litres of diesel
 7014 litres of petrol
Net margin (diesel + petrol): 13879,56552 CZK
Monthly income: 416386,9657 CZK
= PROFITABLE FOR EVERY LOAN VARIANT
Remember our fixed costs:
Variant 5.5
139 438,48 + 35 000 = 174 438,48 CZK
Varian 6
81 044,97 + 35 000 = 116 044,97 CZK
Variant 6,5
63 590,84 + 35 000 = 98 590,84 CZK
So as you can see we are getting closer to a state of non profitable environment. In Scenario 3 I am gonna show you the case with most accurate number of customers.
Scenario 3
= THE MOST LIKELY
 average 250 customers per day
 peak 05:00  11:00 + 15:00  17:00
 2 fuel tanks
 low prices
This scenario is example of how it is gonna most likely be after starting our business. Customers wont believe this new technology and the will prefer traditional ways of refueling thier vehicles. The only possible way how to achieve at least a few hundred customers a day is to lower our prices bellow prices of our rivals. Since we have so little customers lets start again with 1 fuel dispenser and I think that we wont need more than 2.
As we can see the solution with one fuel dispenser is not optimal. The are 10 customers average that leave due to anger of waiting too long. We can also think off the efect of seeing 6 cars in the queue. If it was my call I would probably leave. So we need at least one more fuel dispenser.
2 fuel dispensers ( + 400 000 CZK for fuel dispenser and asphalt road)
Now we have 4 customers max waiting in the queue for max 5,5 minutes. That is not indeed bad. Due the fact that we need to keep our costs low we will stick with 2 dispensers. Third one would probably lower the waiting time to 23 minutes but I assume that every potentional customer in this scenario is atracted by low price pro gas and wiling to wait for 5,5 minutes to get his cheap gas.
a) low prices  diesel 35,10 and petrol 34,70
Litres of fuel sold:
 5271 litres of diesel
 3514 litres of petrol
Net margin (diesel + petrol) per day: 6953,634624 CZK
Monthly income: 208609,0387 CZK
= PROFITABLE FOR EVERY LOAN VARIANT
b) attractive price  diesel 34,60 and petrol 34,20
Litres of fuel sold:
 5271 litres of diesel
 3514 litres of petrol
Net margin (diesel + petrol) per day: 2604,690654 CZK
Monthly income: 78140,71961 CZK
= NOT PROFITABLE FOR EVERY LOAN VARIANT
So what is profitable for us, what loan variant we can choose?
We need 6 900 000 CZK for realization of this scenario.
Lets see our fixed costs:
Variant 5.5
131 798,02 + 35 000 = 166 798,02 CZK
Varian 6
76 604,15 + 35 000 = 111 604,15 CZK
Variant 6,5
60 106,41 + 35 000 = 95 106,41 CZK
In this case we have no other option than raise of our prices to level when we earn at least 95 104,41 CZK a month.
Results
I discovered that in any case there will be always 2 fuel tanks. We want to keep our prices low as possible so we would take 20000 litres of gas every time. It is also not good for the business to have only on fuel dispenser. Two would be enough for cases with lower number of customers. I can say that for more than 350 customers per day we would need 3 fuel dispensers to keep our customers satisfied.
Scenario 1 = PROFITABLE
Scenario 2 = PROFITABLE
Scenario 3 = NON PROFITABLE with low prices
As we can see from results of scenario 1 & 2 if we can sustain high number of customers and reasonable prices we can expect great profit from realization of this project. But I am reasonably pessimistic so I think that real number of cutomers per day would be around 250 with low prices of our petrol and diesel. Unfortunately low prices don´t bring us large profit, and we could get to the point where our project is not profitable anymore. From scenario 3 results I know that if I use prices that would be already low for Benzina or other gas stations I can make nice profit. Bud this case in not so easy. Selfservice gas station is a new technology tested and running sufficiently in western countries. So I can expect low confidence of Czech customers and no will to try this technology. For this reason the price must be low asi possible.
So how low could be the price to make at least some profit?
Lets assume that difference between petrol and diesel must be at least 0,40 CZK. That is the situation on gas market. We now that prices 34,60 and 34,20 would make loss of 20 000 CZK a month. Now we can try add 0,10 CZK to prices of petrol and diesel and see if we can make profit.
Diesel: 34,60 CZK
Petrol: 34,20 CZK
Net margin per day: 2604,690654 CZK
Monthly income: 78140,71961 CZK
= NOT PROFITABLE
Diesel: 34,70 CZK
Petrol: 34,30 CZK
Net margin per day: 3474,479448 CZK
Monthly income: 104234,3834 CZK
= PROFITABLE FOR 15y LOAN
Diesel: 34,80 CZK
Petrol: 34,40 CZK
Net margin per day: 4344,268242 CZK
Monthly income: 130328,0472 CZK
= PROFITABLE FOR 15y LOAN and 10y LOAN
Diesel: 34,90 CZK
Petrol: 34,50 CZK
Net margin per day: 5214,057036 CZK
Monthly income: 156421,7111 CZK
= PROFITABLE FOR 15y LOAN and 10y LOAN
Diesel: 35,00 CZK
Petrol: 34,60 CZK
Net margin per day: 6083,84583 CZK
Monthly income: 182515,3749 CZK
= PROFITABLE FOR ALL LOAN VARIANTS
Conclusion
Simprocess proved me that it is a powerful tool to simulate this situation. Thanks to the simulation I was able to set the number od fuel dispenser to right value to keep customer satisfaction. It also showed me aproximately the time for waiting in the queue and simplify my calculations of fuel sold.
I proved that under certain circumstances this project is well realizable. But there is so many factors that can be added to simulation or project itself. For example selling a destiled water or washer mixure. But to calculate every possibility would take days and weeks. I think that this project would be an excellent topic for the bachelor thesis but with every possibility it is just too big for purposes of this course.
Would I start this project with the information I have gained by simulation?
No I probably wont. The simulation provided me one incredibly valuable information. And that was the whole reason of this simulation. This project is under certain circumstances profitable and it shloud be considered for future business. But for doing this I would need more information, oppinions of professionals, business plan and permits and reports from various local authorities. Than I would make more calculations, more complex simulations in SimProcess and than I would start the business.