Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 13

Rehan Ali Mardhani

Mathematic Analysis and Approaches HL

Optimizing the profits for my business - Amigo

Number of Pages: 13

1
Rehan Ali Mardhani

Introduction

During the 2021 summer holidays, I started my own business and called it Amigo. I started

selling artificial plants on Jumia, an e-commerce platform, as well as through sales agents, whom

I pay 10 % commission of the total sales they make. For the first few months, I bought these

artificial plants from my supplier, only when I got an order. I update my financial reports at the

end of every month, and after a few months, I realised that buying them together in the first week

of each month would save me transport costs, and would increase my profits. Thus, I decided to

buy a fixed quantity of artificial plants each month. However, I soon realized not all of the plants

I bought were sold, leading to higher costs and reduced profits. Therefore, I decided to

investigate how many artificial plants I need to buy each month, such that all the products are

sold and the profits are maximized. During my maths classes, I was introduced to optimization,

in calculus, and I will be using this method for this investigation.

Aim

The aim of this investigation is to find out the units of artificial plants I should buy for my

business (Amigo) in order to optimize profits.

Method

To achieve my aim, I will need a Profit equation in terms of units bought, P ( x ) , where x

represents the units bought. And to get a profit equation, I will have to subtract the cost equation

from the revenue equation. Therefore, P ( x ) =R ( x ) −C ( x ) Whereby, P(x ) is the profit equation,

R(x ) is the revenue equation, C ( x) is the cost equation, and x is the units bought. I will be using

polynomial regression to model the revenue and cost equations. Then subtract the revenue

2
Rehan Ali Mardhani

equation from the cost equation to get the profit equation. Then, I will use first and second

differentiation to find out the maximum and minimum points. The maximum point will represent

the number of artificial plants I should buy for my business (Amigo) to optimize profits. I will

then substitute this value into the profit equation P( x ), to determine the profit my business will

earn each month after selling the products.

Exploration

Table 1 shows a summary of the units bought, sold, total revenue, costs and profits of Amigo for

the months June 2021 to May 2022.

Table 1

A total revenue and a total cost equation will be required in order to get the total profit equation.

Therefore, I will start by modelling an equation for the total revenue.

1. Modelling the Revenue Equation

Since I was not able to sell all the products I bought in many months, using the equation

R ( x )= price× quantity sold is not plausible. When I graphed the revenue and the Units bought

values, I realized that the plots are spread out, and using polynomial regression will be more

appropriate that using linear regression. Therefore, I decided to use polynomial regression to

3
Rehan Ali Mardhani

model the revenue equation. Polynomial regression is a form of regression analysis in which data

is fit to a model and then expressed as a mathematical function. Polynomial regression relates the

two variables
350,000
(x and y) in a
300,000
curved
250,000
Total Revenue (UGX)

200,000

150,000

100,000

50,000

-
2 3 4 5 6 7 8 9 10 11
Units Bought
relationship. (KENTON, 2022)

Graph 1

4
Rehan Ali Mardhani

After plotting the points on the graph, I assumed that the R-squared value I will obtain will be

low, and the co-efficient values will be inaccurate. Therefore, I decided to concatenate and make

the data for each quarter of my business’s financial year (June 2021 – May 2021). Therefore,

 June-2021, July-2021 and August-2021 (Q1)

 September-2021, October-2021 and November-2021 (Q2)

 December-2021, January-2022 and February-2022(Q3)

 800,000 March-2022,
700,000
April-2022
600,000
Total Revenue (UGX)

and May-
500,000

400,000 2022(Q4)
300,000

200,000
Table 2 100,000
shows the

data for -
12 14 16 18 20 22 24 26 28
each quarter.

Graph 2, Units Bought shows the

points plotted.

Table 2

5
Rehan Ali Mardhani

Graph 2

Since Graph 2 looks more appropriate, and will give me a higher R-squared value, I will be using

data from Table 2 for modelling the cost equation.

Then I performed Quadratic and cubic regression using my Ti-84 Plus CE calculator, by pressing

STAT, chose Edit, and I entered the units bought in L1, and revenue in L2, where L is short for

List. Then I again pressed STAT, chose CALC, QuadReg, and entered L1 for Xlist, and L2 in

Ylist. I repeated these steps for CubicReg as well. I performed these steps for both the monthly,

and quarterly values. As I expected, the monthly values gave me a lower R-squared value, while

quarterly values gave me a higher R-squared value. R-squared (R2) is a statistical measure that

represents the proportion of the variance for a dependent variable that's explained by an

independent variable or variables in a regression model (Fernando, 2021). Figure 1 and 2 show

the results for monthly values. Figure 3 and 4 show the results for Quarterly values.

Figure 1 Figure 2

6
Rehan Ali Mardhani

Revenue
800,000
700,000
600,000
Total Revenue (UGX)

500,000
400,000
300,000
200,000
100,000
-
12 14 16 18 20 22 24 26 28
Units Bought

Figure 3 Figure 4

Cubic regression of the quarterly values gives R-squared value to exactly 1. Which means that

the values are 100% accurate. Hence, I will be using Cubic regression of quarterly values (figure

4). Therefore, my Revenue equation is

R ( x )=−2242.424242 x 3 +136212.1212 x 2−2661848.485 x +17141515.15

Graph 3 shows the graph of the above revenue equation. And the dots represent the values of

units bought, and revenue from table 2.

Graph 3

2. Modelling the Cost Equation

I will using polynomial regression to get a cost equation as well. By following the steps, I used to

model the revenue equation, I got the following results. Figure 5 and 6 show the results for

7
Rehan Ali Mardhani

monthly quadratic and cubic regression. Figure 7 and

8 show the results for quarterly quadratic and cubic

regression.

Figure 5 Figure 6

Figure 7 Figure 8

Cubic regression of the quarterly values gives R-

squared value to exactly 1. Which means that the

values are 100% accurate. Hence, I will be using

Cubic regression of quarterly values (figure 8).

Therefore, my Cost equation is

C ( x )=112.1212121 x3 −5143.939394 x2 +78092.42425 x−133742.4243

8
Rehan Ali Mardhani

Graph 4 shows the graph of the above revenue equation. And the dots represent the values of

units bought, and revenue from table 2.

Graph 4

3. Modelling the Profit Equation

To get the Cost


Profit 450,000
400,000
equation, I 350,000 will
Total Revenue (UGX)

300,000
subtract 250,000 the
200,000
cost
150,000

equation 100,000
50,000
C ( x ) from - the
12 14 16 18 20 22 24 26 28

revenue Units Bought

equation R( x )

9
Rehan Ali Mardhani

P ( x ) =R ( x ) −C ( x )

P ( x ) =(−2242.424242 x 3 +136212.1212 x 2−2661848.485 x +17141515.15 )−(112.1212121 x 3−5143.939394 x 2 +7

P ( x ) =−2242.424242 x 3 +136212.1212 x 2−2661848.485 x+ 17141515.15−112.1212121 x 3 +5143.939394 x 2−780

3 2
P ( x ) =−2354.54 5454 x +141356.0606 x −2739940.909 x +17275257.57

Now that I have the profit equation, I will proceed by using calculus to find the units Amigo

should buy to maximize its profits.

4. Profit Optimization

Now, I will get the first differention of P( x ) and then equate it to 0.

3 2
P ( x ) =−2354.54 5454 x +141356.0606 x −2739940.909 x +17275257.57

P ' ( x )=3 ¿

P ' ( x )=3 ¿

P' ( x )=−7063.636362 x 2+ 282712.1212 x−2739940..909

2
−7063.636362 x +282712.1212 x−2739940..909=0

10
Rehan Ali Mardhani

Now I will solve this equation, and get the values of x . To do this, I pressed apps on my TI-84

Plus CE calculator, chose PlySmlt12, Polynomial Root Finder, Order: 2. Then I entered the

values from the above equation into the appropriate field, and chose Solve. Figure 9 shows the

results.

Figure 9

Therefore, the two x values we got are:

x 1=16.46532926

x 2=23.55837577

Now I will proceed with second differention.

Then I will substitute the x values we got into the


''
P ( x ) to determine which one is maximum and minimum.

' 2
P ( x )=−7063.636362 x + 282712.1212 x−2739940..909

2−1 1−1
P ' '(x )=−2 ( 7063.636362 ) x +1 ( 282712.1212 ) x

''
P ( x )=−14127.27272 x +282712.1212

11
Rehan Ali Mardhani

''
P ( x 1 )=−14127.27272 x1 +282712.1212

P' ' ( 16.46532936 )=−14127.27272(16.46532926)+282712.1212

¿ 50103.47832

As the above value is greater than 0 (>0), this is the minimum point.

''
P ( x 2 )=−14127.27272 x 2+282712.1212

P' ' ( 23.55837577 )=−14127.27272(23.55837577)+282712.1212

¿−50103.47814

As this value is less than 0 (<0), this is the maximum point. Therefore, 23.55837577 is maximum

units Amigo should buy each quarter. As 0.55837577 unit cannot be bought, I will replace the x

value in the original profit equation P( x ) with 23 and 24. The value obtained will be the profits I

will earn after selling x units. The greater value will therefore mean that it’s the highest profit I

can make. The value (23 or 24) that gives a higher profit, will be number of units I should buy

each quarter for my business to earn the maximum profit.

P ( x ) =−2354.54 5454 x 3 +141356.0606 x 2−2739940.909 x +17275257.57

3 2
P ( 23 )=−2354.54 5454 ( 23 ) + 141356.0606 ( 23 ) −2739940.909(23)+ 17275257.57

¿ 386218.1816

¿ 386218

12
Rehan Ali Mardhani

3 2
P ( 24 ) =−2354.54 5454 ( 24 ) +141356.0606 ( 24 ) −2739940.909(24)+17275257.57

¿ 388530.3035

¿ 388530

Since buying 24 units gives a higher profit than 23 units, I should buy 24 units of artificial flower

each quarter to maximize its profit. However, my aim was to identify how many units Amigo

should buy each month. Therefore:

24
=8
3

388530
=129510
3

I should buy 8 artificial flower each month, which will give me a profit of approximately

129,500 UGX.

Bibliography
Fernando, J. (2021, 09 12). investopedia. Retrieved from investopedia:
https://www.investopedia.com/terms/r/r-squared.asp

KENTON, W. (2022, 05 29). investopedia. Retrieved from investopedia:


https://www.investopedia.com/terms/n/nonlinear-regression.asp#citation-1

13

You might also like