Chapter 3 'Forecasts'

Southwestern University
Southwestern University (SWU) is a large public university located in Stephenville, Texas, with a
enrollment of nearly 20,000 students. The school has a dominant presence in the social environment of this
small town, since during the autumn and spring there are more students than permanent residents.

Always recognized as a powerhouse in American football, the SWU generally ranks among the top 20.
of college football. Since hiring the legendary Phil Flamm as head coach in 2006 (with the
Hope of reaching the desired number 1) increased attendance on the five Saturdays of home games.
Before Flamm's arrival, the average attendance was generally between 25,000 and 29,000 people per game.
But with just the announcement of their arrival, ticket sales for the season increased by more than 10,000.
Aficionados. Stephenville and SWU were ready for great things!

However, the immediate problem facing the SWU was not its position in the NCAA; it was the
capacity of its stadium. The current SWU stadium, built in 1953, had a capacity for 54,000
followers. The table below indicates attendance at each game over the last 6 years.

One of Flamm's demands since joining the SWU has been the expansion of the stadium or even the
construction of a new one. With the increase in attendance, the SWU administrators began to see
the way to solve the problem. For Flamm, the individual dormitories of his athletes had to be part of
of any expansion.

The president of SWU, Dr. Joel Wisner, decided it was time for his vice president of development
would predict when the current stadium would be insufficient. In his mind, the expansion was a fact, but Wisner
I should have known how long I could wait. Likewise, I looked for a projection of the income, assuming a
average price per ticket of 50 dollars in 2013 and an annual increase of 5% in future prices.

2007 2008 2009

GAME ASSISTANCE OPPONENT ASSISTANCE OPPONENTS ASSISTANCE OPPONENT
1 34200 Rice 36100 Miami 35900 USC
2a 39800 Texas 40200 Nebraska 46500 Texas Tech
3 38200 Duke 39100 Ohio State 43100 Alaska
4b 26900 Arkansas 25300 Nevada 27900 Arizona
5 35100 TCU 36200 Boise State 39200 Baylor
2010 2011 2012
GAME ASSISTANCE OPPONENT ASSISTANCE OPPONENTS ASSISTANCE OPPONENT
1 41900 Arkansas 42500 Indiana 46900 LSU
2a 46100 Missouri 48200 North Texas 50100 Texas
3 43900 Florida 44200 Texas A&M 45900 South
Florida
4b 30100 Central 33900 Southern 36300 Montana
Florida
5 40500 LSU 47800 Oklahoma 49900 Arizona
State
aWelcome games.
bDuring the fourth week of each season, Stephenville organizes a gigantic crafts festival. This event attracts
tens of thousands of tourists to the town, especially on weekends, and it has a noticeable negative effect on attendance at the games.
Questions for analysis.

1. Develop a forecasting model justifying your selection over any other techniques and project it.
attendance until 2014.

There is a medium-term forecast (2 years)

The trend projection fits a trend line to a series of historical data points and then
project that line into the future to obtain medium and long-term forecasts.

Time series models such as intuitive approach, moving averages, and exponential smoothing are
They are only used for short-term projections, which is not our case.
Linear regression is not recommended to use as we can see in the following figures the trend
These assistances do not follow that pattern over time, neither in general nor by year or by weeks.
of game.

General Assistance
50000

45000

40000

35000

30000

25000
0 5 10 15 20 25 30

Games of Week 1 Games of Week 2

49000 54000
44000
44000
39000
34000 34000
2007 2008 2009 2010 2011 2012 2007 2008 2009 2010 2011 2012

Games of Week 3 Games of week 4

49000 40000
44000 35000
39000 30000
34000 25000
2007 2008 2009 2010 2011 2012 2007 2008 2009 2010 2011 2012

Week 5 Games
55000
45000
35000
25000
2007 2008 2009 2010 2011 2012

Stadium attendance from 2007 to 2012, 5 weeks are played per year.
Now, as we can see in our data, we are presented with a situation of seasonal variation, for
For example, in week 2 is the opening game (maximum attendance) and in week 4 the festival overlaps.
from the town (minimal attendance) so the most advisable thing would be to use a seasonal model
multiplicative, if this seasonal variation is not considered, the magnitude of the seasonal data would be
distorted by the trend.

The multiplicative seasonal model was applied in an Excel spreadsheet.

Now the total annual attendance for the next two years is estimated, for this we see that the annual attendance
if it follows a linear trend, so we can estimate attendance for 2013 and 2014

Total annual attendance

230000

220000
y = 11041x - 2E+07
210000 R² = 0.968
200000

190,000

180000

170000

160000
2006.5 2007 2007.5 2008 2008.5 2009 2009.5 2010 2010.5 2011 2011.5

Series1 Linear (Series1)

2013= (
11041∗ 2013− )
21986800 = 238733 2014= ( 2- 1986800=
11041∗ 2014 ) 249774

Note, we use linear regression only for an estimate of the total annual in the next two years, we are not
considering it as a prediction, it is just an estimate for the calculation of the prediction.

Finally, we divided each estimate of total annual attendance by the number of weeks, after that
we multiply by the seasonal index for that week. This provides the seasonal forecast.
2. What revenues should be expected in 2013 and 2014?

We already have ticket assistance for the next two years. We consider a ticket price for the
2013 of 50 dollars with a 5% increase per year.

3. Analyze the school's alternatives

As we can see, it has been predicted that for the game in week two in 2014, attendance will exceed
the capacity of the current stadium (54,000 people), forecasting an attendance of 56,770 people, this is
It is to be expected since the welcome game takes place that week, so it is not surprising that it is
in this game when the stadium no longer provides supplies.

Now then, considering that we are at the beginning of 2013 or the end of 2012, the SWU would have a period of 1
year for the expansion of the stadium in terms of capacity supply, however, it is important
It should be noted that none of the other games are predicted to exceed the current capacity of the stadium.
so it could extend that period to 2 years, 2 years to be able to execute a renovation of the stadium, with
which would lose about 2700 people from the game.

The school's alternatives are: Renovate in one year and not have any events with insufficient capacity or
remodel in two years but leave 2700 people without attending the welcome game, something that for the people of
The town would not be pleasant at all, besides the fact that the Americans love American football; it's something sacred.
for them, so it's pretty tough :’v.
3.- Forecasts at Hard Rock Café
With the growth of Hard Rock Café, there was also a demand for better forecasts across the
corporation. Hard Rock obtains long-term forecasts to establish a capacity plan, and medium-term
Deadline to close contracts with suppliers. Their short-term sales forecasts are made every month,
for coffee, and then they are added to have a general overview in the head office.

The heart of the sales forecasting system is the point of sale system, which captures data from the
transactions. The sale of each item represents a customer; the sales data of the items is
They transmit every day to the database of the corporate offices in Orlando. There, the finance team
the forecasting process begins, forecast the monthly account of customers, retail sales, sales of
banquets and concert sales for each café.

The general managers of each café connect to the same database to prepare the daily forecast.
From its establishment. A restaurant manager obtains information about the sales from that day the previous year,
adding local data on upcoming events. The daily sales forecast is broken down by hour, the
cual sirve de guía para la programación del personal. Los paquetes de programación computarizada asignan a
people to their workplace. The variations between forecasts and actual sales are examined
afterwards to observe why the errors occurred.

An even more sophisticated application of statistics can be found in the menu planning of Hard Rock.
Through the use of multiple regression, managers calculate the effect on the demand for other items of
menu when the price of one of them changes. For example, if the price of the cheeseburger
increase, Hard Rock predicts the effect this will have on sandwich sales and salads.
The administrators perform the same analysis with the menu distribution, where the central section generates
higher sales volumes.

Analysis
Hard Rock Café uses a forecasting system to have control and a basis for its decision-making.
These forecasts help them support with truthful information the changes they are going to make regarding the
last year and they promise to reduce losses. Currently, forecasts are used to understand how
the price change in some of its products will affect the sales of the others. Likewise, it
use to have a correct distribution of the menu, and depending on the section in which it is located
consider how this will affect the sale of other dishes. The scheduling of the staff is also
broken down based on the number of hourly sales forecasted taking into account the database and
the events that will happen on those dates.

Other applications that can be made include knowing which areas of the restaurant spend the most, and that
type of tables, as well as the number of people in order to have a arrangement for the waitstaff
optimal, where the entire workload is not placed on a few and there is mobility when attending to tasks
to the people. The number of times maintenance is performed on things should be an indicator of when
something is no longer working and needs to be replaced. The number of people dedicated to cleaning the place
It should be proportional to the number of people in the café to ensure that the place always looks good.
impeccable.

The point of sale system is the basis for conducting the statistical study required to make the
forecasts, it is the way to account for the sales of each item, know at what time sales were made
And what hours are expected to have more customers and therefore will require more staff, which employee has
a better sales number, etc. This is why their role is crucial when making a correct
analysis of each coffee.

It is necessary to take into account the previous years because one year is not a sufficient indicator; it is possible that in that
year there were events that gradually increased sales or on the contrary that political conflicts or
global economic factors affect sales and this does not necessarily define the efficiency of a good manager and the
amount of bonuses that he deserves for his productivity. A longer time range leads you not to neglect
ningún año de trabajo y te evalúa de una mejor manera tu trabajo de una manera global pues tu meta no es
only a good year of sales.
The political aspect is not being considered; it is possible that if there are problems in a country, or if it is an election year
elections, the amount of money that families spend on eating out decreases, and this affects the number
of sales, similarly this can impact the devaluation of a currency, or in an economic crisis.
These situations that society is facing are not being taken into account.

Regression Exercise
En el restaurante Hard Rock de Moscú, el gerente está tratando de evaluar cómo afecta una nueva campaña
from advertising to the clients' accounts. Using the data recorded for the last 10 months, develop
a least squares regression relationship and then forecast the expected number of customers
when the advertising is 65,000 dollars.

Regression between Clients vs Advertising

70000

60000

50000

40000

30000

20000

10000

0
14000 17000 25000 35000 45000 50000 60000
Advertising

The equation that gives us the regression is:

( = 0.79965 * Publiidad+
) 833.63

Therefore, with an advertising budget of 65,000 dollars, 60,307 customers are expected.
 AUTONOMOUS UNIVERSITY OF
NEW LEON

College of Chemical Sciences

Chemical Engineering

Operations Management
Case Studies: Chapter 3

Facilitador:Pedro Pablo Tamez Vargas

Semester: 9th
Team:
Sepúlveda González María Fernanda 1619203
Torres Juárez Fernando Hans 1592369
Roberto Tadeo Vidales Zúñiga 1640140
Jacobo Contreras Albed Francisco 1596868

Cd. Universitaria, San Nicolás de los Garza, Nuevo León on the 9th of
February 2018