We can first use descriptive statistics by examining historical data on customer flow

Examine the number of customers based on different days of the week, month (perhaps even year).
Also examine the number of customers at different times of the day.
By summarizing the common traits of busy times/days, we can develop a strategy on when to open up more cash
The most important data would be the number of customers per time period (hour, or 15-minutes, depending on

even year).

hen to open up more cash registers.

5-minutes, depending on how flexible the work force scheduling is).

Arrival Day of The Week (Month/Year)

Length of Stay
Use of Mini Bar
Cash or Credit Customer
Use of Extra Hotel Services (such as Wi-Fi, Room Service, On Demand Movies etc.)

Using these data measurements, the hotel can decide which customers are likely to spend more money within the
For example, if a certain group of customers are likely to spend a lot of money on room service, those customers m
Or, using the arrival day and length of stay, one can identify whether a customer is a business traveller or not (we
These business travellers might spend more money if their company is paying for the trip and identifying them ma

nd more money within the hotel.

service, those customers may be offered discounts at nightly stay prices.
iness traveller or not (we can assume that they tend to arrive weekdays and don't stay the weekend)
p and identifying them may prove to be lucrative.

Customer Arrival Date

Customer Arrival Time
Customer Service Time
Purchase Type
Revenue Generated

Just using these basic types of data, a fast food restaurant will be able to identify rush hours in a given day.
Using this information, they can decide how many registers to open at different times of the day.
Also looking at the purchase patterns, they can decide on how to stock up on different food items on different day

ours in a given day.

ood items on different days of the week.

Cust ID

Region

10001 East
Ordinal

Categorical

Payment

Transaction Code
Source

Paypal

93816545

Web

Categorical

Ordinal

Categorical

Amount

Product

$20.19 DVD
Ratio

Categorical

Time Of Day

22:19
Interval

Homeowner
Y

Categorical

Credit Score
Years of Credit History
Revolving Balance
Revolving Utilization Decision
725
20
$
11,320
25%
Approve

Interval

Interval

Ratio

Ratio

Categorical

Gender
Categorical

Age
Interval

Ethnicity
Categorical

Length of
Residency
Interval

Quality of
Satisfaction
Schools
Ordinal
Ordinal

MODEL:
BALANCE = -17,732 + 367 x AGE + 1300 x YEARS EDUCATION + 0.116 x HOUSEHOLD WEALTH
a.

b.

367

The average account balance increases by approximately $367 for each year

1300

The average account balance increases by approximately $1300 for each yea

0.116

The average account balance increases by approximately $0.116 for each $1

AGE
EDUCATION

36 years old
16 years

WEALTH $ 175,000.00
PREDICTED BALANCE $ 36,580.00

pproximately $367 for each year increase in AGE

pproximately $1300 for each year increase in EDUCATION

pproximately $0.116 for each $1 increase in WEALTH

MODEL:
D = k - pP + aA + tT + qQ
a.

P:

As Price increases, Demand goes down.

A:

As Advertising increases, Demand goes up.

T:

As Transportation increases, Demand goes up.

Q:

As Product Quality increases, Demand goes up.

b.

The variables do not influence each other.

c.

The relationship of D to P is overly simplistic. If P is too high, the model predicts negative D, in fact D w
The variables might influence each other as well. For example, high production quality may cost more a

cts negative D, in fact D will be at least ZERO.

on quality may cost more and hence may have a higher price tag.

MANUFACTURE
Variable Cost $
9.00 /unit
Fixed Cost $ 4,000.00

a.

b.

OUTSOURCE
Variable Cost $ 12.00 /unit
Fixed Cost $
-

VOLUME
1000 units
Cost of Manufacturing $ 13,000.00
Cost of Outsourcing $ 12,000.00 <----

BETTER OPTION

BREAK EVEN FOR MANUFACTURING

at least

444.44444 units

BREAK EVEN FOR OUTSOURCING

at least

0 units

INDIFFERENCE POINT
1,333.33 units
which means
If VOLUME is less than or equal to 1333 units, OUTSOURCE
If VOLUME is more than or equal to 1334 units, MANUFACTURE

E: Earnings
T: Turnover
S: Sales
C: Cost of Sales
TI: Total Investment
CA: Current Assets
FA: Fixed Assets
MC: Mill Cost of Sales
SC: Sales Expense
FC: Freight and Delivery
AC: Admin Costs
TI = CA + FA

Current Assets

T = S / TI

Turnover

Total Investment

Fixed Assets

ROI = T * E / S

ROI

Earnings

Turnover

SALES

E=S-C

Cost of Sales

Mill Cost of
Sales

Selling
Expense

Freight &
Delivery

C = MC + SC + FC + AC

Freight &
Delivery

Admin
Costs

10

b
x
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2.25
2.50

-0.25
14.14
11.89
10.75
10.00
9.46
9.04
8.69
8.41
8.16
7.95

0
10.00
10.00
10.00
10.00
10.00
10.00
10.00
10.00
10.00
10.00

0.5
5.00
7.07
8.66
10.00
11.18
12.25
13.23
14.14
15.00
15.81

1
2.50
5.00
7.50
10.00
12.50
15.00
17.50
20.00
22.50
25.00

When b<0, more marketing effort means less demand

When b=0, marketing effort has no effect on demand
When 0<b<1, we can observe law of diminishing returns
When b=1, the relation between marketing effort and demand is linear
When b>1, the returns in demand increase for every additional unit marketing effort
The models assume that we can clearly measure marketing effect and demand.
Also, we do not consider any other factors that affect the demand.
Intuitively, the models when b<0 and b=0 and b>1 does not make much sense.
It is likely that 0 < b <= 1
To determine a good estimate for b, we need to collect data and fit it to the given curve.

1.5
1.25
3.54
6.50
10.00
13.98
18.37
23.15
28.28
33.75
39.53

SAMPLE SKETCHES
45.00
40.00
35.00
30.00
b<0
25.00

b=0
0<b<1

20.00

b=1
b>1

15.00
10.00
5.00
-

0.50

1.00

1.50

2.00

2.50

3.00

MODEL:
G = (m x d ) / vf
m
d

24 miles
20 days
480 total miles per month

vf

30 mpg

16 gallons used per month

DEMAND MODEL
D = 2000 - 3P
COST MODEL
C = 5000 + 4D = 5000 + 4 x ( 2000 - 3P) = 13000 - 12P
TOTAL REVENUE
TR = D x P = ( 2000 - 3P) x P = 2000P - 3 P^2
TOTAL COST
TC = 13000 - 12P
TOTAL PROFIT
TP = TR - TC
= 2000P - 3 P^2 - (13000 - 12P)
= -13,000 + 2012 P - 3 P^2

$
$

P
600.00
300.00

Revenue

D
500
1200

=PxD
= P x (-2.333 P + 1900 )
= -2.333 P^2 + 1900 P

Price vs Demand
1400
1200
1000
800

D = -2.333 P + 1900

600
400
200

TRIAL AND ERROR

P
Revenue
$
1.00 $
1,898
$
10.00 $ 18,767
$ 100.00 $ 166,670
$ 1,000.00 $ (433,000)
$ 500.00 $ 366,750
$ 750.00 $ 112,688
$ 250.00 $ 329,188
$ 375.00 $ 384,422
$ 425.00 $ 386,102 <---$ 450.00 $ 382,568

0
$-

BEST GUESS SO FAR

$100.00

$200.00

$300.00

Price vs Demand

2.333 P + 1900

$300.00

$400.00

$500.00

$600.00

$700.00