Analysis and Recommendations for Tully Tyres

The assignment content involves a Monte Carlo simulation model to examine the next 12 months' average monthly profit of Tully Tyres, with changes in average selling price and profit margin. The model is tested for goodness of fit using regression analysis, and Model 2 (dependent variable: over-head cost, independent variables: batches) is found to be the most suitable. Additionally, the assignment also involves calculating break-even points for two products A and B, determining total targeted sale volume, and calculating expected profit before and after tax.

DECISION SUPPORT TOOLS
DECISION SUPPORT TOOLS
Assessment item - 3
Student name and id
[Pick the date]

Question 1
(a) There are differences between decision making under certainty, under risk and complete
uncertainty. Decision under certainty may be defined as the situation in which the decision
making authority has complete information in relation to the decision and thus evaluation of
alternatives and choosing the right one becomes easy. Usually these decisions are taken
based on already framed rules since these situations are recurring (Holsapple, & Whinston,
2013).
In case of decision under risk, only limited information tends to be available with the decision
making outcome. Hence, subjective probabilities of the possible outcomes need to be computed
and then through various decision making tools the alternative that leads to the maximum value
creation is chosen. In case of decision under complete uncertainty, the relevant decision maker
possesses no information in relation to the various possible scenarios and respective chances of
these happening. Hence, decision making through convention tools or wisdom is quite difficult
which is why creative thinking is deployed on the part of the decision maker (Hair, et.al., 2015)
b) 1) Based on the following table, the optimist would prefer the share market since it can give
the maximum possible returns.
2) Based on the following table, the pessimist would prefer the share market since it can give the
maximum returns in the worst possible case.
1

3) Based on the following regret matrix, it is apparent that real estate ought to be selected.
4) We need to compute the EMV or Expected Monetary Value for all the given choices.
Stock Market (EMV) = 0.3*80000 + 0.7*(-20000) = $ 10,000
Bonds (EMV) = 0.3*30000 + 0.7*20000 = $ 23,000
Real estate (EMV) = 0.3*25000 + 0.7*15000 = $ 18,000
Based on the above, it is apparent that investment should be made in bonds.
5) The relevant formula to be used is highlighted below.
2

Thus, based on the given information provided and the above formula, EVPI has been computed
below.
Question 2
(a) In order to decide on the decision which Jerry should make, the following table would prove
useful.
The respective EMV computation is shown below.
Large Bicycle Shop (EMV) = 80000*0.5 + (-40000)*0.5 = $ 20,000
Small Bicycle Shop (EMV) = 30000*0.5 + (-10000)*0.5 = $ 10,000
Since, the large bicycle shop would result in higher EMV, hence it would be preferred by Jerry.
(b) The revision of prior probability is based on the following statistics regarding the accuracy of
the market research conducted by the friend.
3

The revision of probability under a favourable study is captured in the following table.
The revision of probability under a unfavourable study is captured in the following table.
c) Based on the table indicated below, the posterior probability may be estimated.
4

It is apparent from the computation carried out that the requisite posterior probability comes out
as 0.25.
d) The market research utility needs to be analysed considering two circumstance namely study
being favourable and unfavourable.
Case 1: Favourable Outcome
Case 2: Unfavourable Outcome
Hence, based on the above two table in the possible scenarios, expected utility = 40,400*0.6 +
0*0.4 = $ 24,240
5

Hence, the value of the information = 24,240 – 20,000 = $ 4,240
Also, it is known that the fee paid to the friend for conducting market research amounts to $
3,000.
Therefore, gains reaped by Jerry by opting for the market research = 4240 – 3000 = $1,240
Considering the above gains, market research should be opted by Jerry.
Question 3
Monte Carlo Simulation
(a) Model from Monte Carlo Simulation to examine the next 12 month’s average monthly
profit.
6

Formula view of model:
(b) Average monthly profit for Tully Tyres for the 12 months period = $962.92 (from
Monte Carlo Simulation model)
(c) The two data variables has been changed and the new model is highlighted below:
Change 1: Average selling price has increased (From $80 to $100)
Change 2: Profit margin has increased (From 22% to 32%)
7

Formula view of model:
8

Date: September 20, 2017
To: Manager, Tully Tyres
Dear Sir
As suggested that the average price would be increased by $ 20, alterations have been made in
the financial model which clearly indicate a surge in the average profit and also the overall
profit.While this is definitely impressive and as per expectations, it is essential to test the core
assumption underlying the above analysis, in relation to the sales being constant.
It may so happen that in a bid to increase the price, the sales of the company may go down which
needs to be guarded against. As a result, conducting a small pilot to understand the impact on
sales by changing the price needs to be undertaken. Further, once implemented also, then also
careful vigilance must be kept on the monthly sales volume so as to identify any downfall in
volume of sales. Appropriate action would need to be undertaken on the basis of customer
response to the increase product price.
Yours Sincerely
STUDENT NAME
Question 4
(a) High – low method
The aim is to calculate the overhead cost based on the factor machine hours. Further, to find the
overhead cost when the machine hours is 3000.
The total overhead cost is combination of fixed cost, per unit variable cost and machine hours.
 Variables cost per unit
9

Highest machine hours = 3800
Lowest machine hours = 1800
Overhead cost @ highest machine hours = 48,000
Overhead cost @ lowest machine hours = 46,000
Variables cost per unit = ( 48000−46000
3800−1800 )=$ 1
 Fixed cost = { 48000− ( 1∗3800 ) }=$ 44,200
Now,
The regression equation for the computation of over-head cost is shown below:
Over-head cost = Fixed cost + (Variables cost per unit * machine hours)
¿ 44200+ ( 1∗3000 )=$ 47,200
Therefore, the over-head cost is $47,200.
(b) Regression model for the overhead cost
Model 1: Depended variable is overhead cost and machine hours as independent variable
10

¿ head cost= ( 59198.78 ) − { ( 2.30 )∗Machine hour s }
R square = 0.0109 (low)
Adjusted R square = -0.1128 (negative value – significantly low)
The low value indicates that the model is statistically not a good fit for analysis.
Assuming alpha as 5%, it would be fair to conclude that p value for MH (machine hours) is 7.77
and this value is significantly higher than alpha. Hence, the slope is not significant and can be
said equal to zero. Further, from ANOVA table, the F statistics is 0.0878 and the corresponding p
value is 0.7744, which is also higher than alpha. Therefore, it can be said that slope of machine
hours can be said equal to zero and hence, the model is not good fit (Holsapple, & Whinston,
2013).
Model 2: Depended variable is overhead cost and batches as independent variable.
11

¿ head cost= ( 6555.56 )− { ( 234.57 )∗Batches }
R square (R2) = 0.8313 (reasonably high)
Adjusted R square (R2) = 0.8102 (reasonably high)
Assuming alpha as 5%, it would be fair to conclude that p value for batches is zero and this
value is lower than alpha. Hence, the slope is significant. Further, from ANOVA table, the F
statistics is 39.42 and the corresponding p value is 0.0002, which is also lower than alpha.
Therefore, the slope of batches is significant and regression model is good fit for analysis (Taylor
& Cihon, 2004).
Model 3: Depended variable is overhead cost and batches and machine hours as independent
variable
12

¿−head cost= [ ( 9205.66 )− { ( 0.93 )∗Machine hours ) }+{(233.83)∗Batches¿ }¿
R square (R2) = 0.8331 (reasonably high)
Adjusted R square (R2) = 0.7854 (high)
Assuming alpha as 5%, it would be fair to conclude that p value for batches is zero and this
value is lower than alpha. Hence, the slope is significant. However, p value for machine hours is
0.79 which is higher than alpha and hence, the slope is not significant. Further, from ANOVA
table, the F statistics is 17.46 and the corresponding p value is 0.001, which is also lower than
alpha. Therefore, the model is good fit for analysis (Eriksson, & Kovalainen, 2015).
(c) The most suitable regression model would be one which has the highest value of
adjusted r square. Even though R square could also been compared but the same is not
done since as the predictor variables increase, the R square tends to increase while
adjusted R square is independent of the predictor count. Thus, the simple regression
model based on batches as the independent variable would be preferred model.
(d) The selected regression model is “model 2.”
13

¿ head cost= ( 6555.56 ) − { ( 234.57 ) ∗Batches }
Total number of batches = 150 (given)
¿ head cost= ( 6555.56 ) − { ( 234.57 ) ∗150 }=$ 41,740.71
Therefore, the over-head cost for 150 batches is $ 41,740.71.
Question 5
(a) The value of unit contribution margin for product A and B
Product Sales price per unit Variable cost per unit Unit contribution margin
A $10 $5 = $10-$5 = $5
B $20 $12 = $20-$12 = $8
(b) The total number of units of B needs to sell at break –even point
Fixed cost for B $4000
Unit contribution margin for B $8
Units needs to be sold at break-even point
¿ ¿ cost for B
Unit contribution margin for B
=$4000/$8 = 500 units
(c) The total number of units of A needs to sell at break –even point
Fixed cost for A $4000
14

Unit contribution margin for A $5
Units needs to be sold at break-even point
¿ ¿ cost for A
Unit contribution margin for A
=$4000/$8 = 800 units
The value of break-even sale volume (for the respective
months in dollars)
= 800* 10= $8000
(d) Ratio of production of A to B is given as 2:1
(i) Total number of units need to produces for product A and B
Expected profit (before tax) $5000
Average contribution margin
{( 2
3 )∗5 }+ {( 1
3 )∗8 }=$ 6
Total targeted sale volume
= (Expected profit +fixed cost)/ Average contribution margin
¿ ( 5000+4000 )
6
¿ 1500 units
Number of units produced for product A (1500)∗( 2
3 )=1000 units
Number of units produced for product B (1500)∗( 1
3 )=5 00 units
(ii) Total number of units need to produces for product A and B
Expected profit (post tax) $21,000
15

Monthly profit before the tax (x)
x (1−30 % ) =21000
$ 30,000
Average contribution margin
{( 2
3 )∗5 }+ {( 1
3 )∗8 }=$ 6
Total targeted sale volume
= (profit before the tax +fixed cost)/ Average contribution
margin
¿ ( 30 000+ 4000 )
6
¿ 5667 units
Number of units produced for product A (5667)∗( 2
3 )=3778 units
Number of units produced for product B (5667)∗( 1
3 )=1889 units
16

Reference
Eriksson, P. & Kovalainen, A. (2015). Quantitative methods in business research (3rd ed.).
London: Sage Publications.
Hair, J. F., Wolfinbarger, M., Money, A. H., Samouel, P., & Page, M. J. (2015). Essentials of
business research methods (2nd ed.). New York: Routledge.
Holsapple, C. & Whinston, B.A. (2013) Decision Support Systems: Theory and Application (6th
ed.). Sydney: Springer Science & Business Media.
Taylor, K. J. & Cihon, C. (2004). Statistical Techniques for Data Analysis (2nd ed.). Melbourne:
CRC Press.
17