ICOTS 2, 1986: Beth H.

Bryan

USING SIMULATION T O MODEL REAL WORLD PROBLEMS

Beth H. Bryan
Augusta College
Georgia, USA

Introduction

T h e p r i m a r y source o f t h e material used i n t h i s presentation i s - T-h e A-r t

a n d ~ e c h n i q u e s o f Simulation, a book f r o m t h e Q u a n t i t a t i v e L i t e r a c y
Series. T h e series was w r i t t e n bv members o f t h e American Statistical A s -
sociation-National Council o f ~ e i c h e r so f Mathematics J o i n t Committee o n
t h e C u r r i c u l u m in Statistics a n d P r o b a b i l i t y a n d f u n d e d in p a r t by a g r a n t
f r o m t h e National Science Foundation. These techniques a r e designed f o r
use in middle school t h r o u g h senior high school. T h e y f e a t u r e statistical
t o p i c s t h a t a r e i m p o r t a n t t o students, a wealth o f hands-on activities, real
data sets a n d a c t i v e experiments w h i c h motivate s t u d e n t participation, a n d
g r a p h i c a l methods instead o f complicated formulas o r a b s t r a c t mathematical
. concepts. I n p a r t i c u l a r , simulation i s i n t r o d u c e d as a t e c h n i q u e f o r s o l v i n g
p r o b a b i l i t y a n d s t a t i s t i c s problems.

Objectives

Practical problems f r o m t h e v e r y simple t o t h e most complex can b e solved

( o r a t least approximated) by u s i n g simple simulation. T h e simulation p r o -
c e d u r e i n v o l v e s c o n d u c t i n g experiments w h i c h closely resemble an actual
s i t u a t i o n i n o r d e r t o p r o v i d e answers t o r e a l l i f e problems.

A hands-on approach i n w h i c h t h e s t u d e n t s a c t i v e l y p a r t i c i p a t e i n o b t a i n -
ing r e s u l t s i s used. B e g i n n i n g w i t h simple models where emphasis i s placed
o n estimating t h e p r o b a b i l i t y o f an event, a g r a d u a l p r o g r e s s i o n e v e n t u a l l y
leads t h e s t u d e n t s t o t h e p o i n t where t h e y are able t o a p p l y methods i n -
v o l v i n g s t a t i s t i c a l i n f e r e n c e t o d r a w conclusions about t h e i r r e s u l t s . S t u -
d e n t s a r e t a u g h t a s t e p - b y - s t e p p r o c e d u r e f o r a p p l y i n g t h e simulation
t e c h n i q u e w h i c h helps t h e m answer questions about t h e b e h a v i o r o f real
processes u n d e r v a r y i n g conditions. Many o f t h e applications discussed
h e r e can r e a d i l y b e adapted t o t h e computer, b u t t h e approach is n o t com-
p u t e r dependent.

A Simulation Model

T h i s method consists o f an e i g h t - s t e p process which i s o u t l i n e d below:

Step 1:

State t h e problem c l e a r l y so t h a t a l l necessary information i s g i v e n a n d t h e

objective i s clear.
ICOTS 2, 1986: Beth H. Bryan

Example: Two e v e n l y matched baseball teams p l a y each o t h e r f o r a series

o f seven games. Estimate t h e p r o b a b i l i t y t h a t team I w i l l w i n t h e series by
w i n n i n g a t least f o u r games f r o m team 2.

Step 2:

Define t h e simple events which f o r m t h e basis o f t h e simulation.

Example: T h e seven games f o r m a series o f seven simple events, each o f

which can b e simulated v e r y easily.

Step 3:

State t h e u n d e r l y i n g assumptions which simplify t h e problem so t h a t a solu-

t i o n can be found.

Example: We make t h e assumption t h a t t h e teams are e v e n l y matched so

t h a t f o r a n y single game (simple event) t h e p r o b a b i l i t y t h a t team 1 w i l l win
so assume t h a t t h e games are independent so t h a t t h e outcome
o f a n y one game i s n o t affected by outcomes o f t h e previous games.

Step 4:

Select a model f o r a simple event by choosing a device t o generate chance

outcomes w i t h t h e probabilities as dictated by t h e real event.

Example: Since t h e p r o b a b i l i t y t h a t team 1 wins a game i s 1/2, ' w e can

model a single game b y tossing a coin and l e t t i n g a head, represent a win
f o r team 1 and a t a i l represent a w i n f o r team 2.

Step 5:

Define and conduct a t r i a l which consists o f a series o f simple event simula-

tions t h a t stop when t h e e v e n t o f i n t e r e s t has been simulated once.

Example: Since teams 1 and 2 are t o p l a y a seven game series, tossing a

f a i r coin seven times would model a single p l a y i n g o f t h e series.

Step 6:

Record t h e observation o f i n t e r e s t by tabulating t h e information necessary

t o reach t h e desired objective. Most often, t h i s simply requires a notation
of favorable o r nonfavorable f o r each t r i a l . Occasionally, a numerical o u t -
come w i l l be noted.

Example: After tossing t h e coin several times, we observe t h e number o f

heads. If t h e number o f heads i s a t least four, t h e t r i a l is classified as
favorable t o t h e event t h a t team 1 wins t h e series. It m i g h t be useful t o
keep a record o f t h e number o f games won by team '1 o n each t r i a l .
ICOTS 2, 1986: Beth H. Bryan

Step 7:

Repeat steps 5 a n d 6 a t least 50 times. A n accurate estimate o f a p r o b a b i l -

ity f r o m empirical r e s u l t s r e q u i r e s a l a r g e number o f t r i a l s . If t h e simula-
t i o n i s done w i t h t h e a i d o f a computer, t h e n 1000 o r more t r i a l s can b e
r u n w i t h o u t a n y inconvenience.

Example: Toss t h e coin seven more times a n d r e c o r d t h e number o f heads.

Repeat t h i s p r o c e d u r e u n t i l a t least 50 t r i a l s of seven coin tosses a r e o b -
tained.

Step 8:

Summarize t h e information a n d d r a w conclusions. We can estimate t h e p r o -

b a b i l i t y o f an e v e n t o f interest, A, by evaluating:

t h e number o f t r i a l s f a v o r a b l e t o A

t h e t o t a l number o f t r i a l s in t h e experiment

Example: We can estimate t h e p r o b a b i l i t y t h a t team 1 w i n s a t least

f o u r games by e v a l u a t i n g :

t h e number o f t r i a l s w i t h a t least f o u r heads

t h e t o t a l number o f t r i a l s in t h e experiment

A coin p r o v i d e d a simple w a y t o generate outcomes in t h e e x p e r i m e n t above

because it was necessary t o use a device t h a t would generate t w o outcomes
w i t h equal f r e q u e n c y . Many o t h e r devices could b e used e q u a l l y as well as
l o n g as t h e r e a r e t w o outcomes w i t h an identical chance o f o c c u r r i n g . F o r
example, we could use a d i e toss a n d c l a s s i f y t h e outcomes as e i t h e r even
o r odd.

Simulation When Probabilities D i f f e r F r o m One-Half

I n t h e p r e c e d i n g example we generated t h e outcomes o f t h e e x p e r i m e n t by

t o s s i n g a coin because each outcome h a d an equal chance o f o c c u r r i n g .
Now l e t u s consider a simulation problem i n which we have a simple e v e n t
w h e r e t h e p r o b a b i l i t y i s n o t equal t o 1/2. Obviously, t h e coin i s n o t an
a p p r o p r i a t e device in t h i s case. However, t h e r e a r e s t i l l a number o f easy
ways t o simulate t h i s t y p e o f experiment. One possibility, i l l u s t r a t e d b e -
low, i s t o use a random number table.

Step 1: State t h e problem c l e a r l y .

Joe r u n s a b u s w i t h e i g h t seats. O n t h e average 10% o f t h e people who

b u y t i c k e t s i n advance d o n o t show u p . So Joe sells t e n t i c k e t s f o r each
trip. Estimate t h e p r o b a b i l i t y t h a t more t h a n e i g h t people show u p w i t h
tickets.
ICOTS 2, 1986: Beth H. Bryan

Step 2: Define the simple events.

The simple events are whether a person holding a t i c k e t shows u p for

the t r i p .

Step 3: State the underlying assumptions.

The probability t h a t any one person w i t h a t i c k e t fails t o show u p for

the t r i p is 10%. Whether any given ticketholder shows u p o r not Is inde-
pendent of what happens t o the other ticketholders.

Step 4: Select a model f o r a simple event.

Draw a number from a random number table. The number 0 will pepre-
sent a ticketholder who didn't show u p f o r the trip.

Step 5: Define and conduct a t r i a l .

Since ten tickets are sold' f o r each t r i p , one t r i a l will consist of d r a p i n g

ten random numbers representing the ticketholders f o r one t r i p . if the
f i r s t t r i a l resulted i n 0 6 4 9 3 1 8 6 6 9 , then nine o f the ten ticket-
holders showed u p f o r the t r i p .

Step 6: Record the observation of interest.

If nine ticketholders showed up, one person did n o t get a seat. We then
keep a tally of the number of ticketholders who did n o t get seats f o r
each t r i p (trial).

Step 7: Repeat step 5 G 6 u n t i l 100 trials are completed.

The results of 100 such trials are summarlxed below:

Number not getting seats Number of t r i a l s

Step 8: Summarize the information and draw conclusions.

The data from step 7 show t h a t more than eight people showed u p 74
times o u t of 100. Therefore the p r o b a b i l j t ~t h a t more than eight ticket-
holders show u p f o r any one t r i p i s estima&$ t o be 74/100 = .74.
ICOTS 2, 1986: Beth H. Bryan

Conclusion

Simulation techniques following the same eight step approach can also be
devised f o r situations where the number of simple events i n a t r i a l is no t
predetermined, i.e. t h e length of a t r i a l changes from one performance t o
the next, as well as more complex problems where t h e event o f interest
may have more than one characteristic. There are numerous examples of
practical problems which can be simulated using this approach and which
are interesting and motivating f o r 12 t o 18 year old students. A partial
listing could include predicting the outcome of sporting events such as
basketball games, t h e results of an election, t h e outcomes of games of
chance, t h e waiting times i n customer lines, and t h e event o f passing o r
failing on multiple choice tests. Clearly, simulation is an ideal mechanism
f o r p r o v i d i n g t h e teacher with the opportunity t o develop a systematic p r o -
gression from estimating probabilities t o drawing conclusions and making
inferences.

References

Gnanadesikan, M., Scheaffer, R. L., & Swift, J. (1986). The a r t and tech-
niques of simulation. Palo Alto, California: Dale Seymour Publications.

Moore, D . S .
(1 979). S t a t i s t i c s , concepts a n d c o n t r o v e r s i e s . San
Francisco, California: W. H. Freeman and Company.

Schools qouncil Project on Statistical Education. (1980). Statistics i n y o u r

world. Slough, Berks: W. Foulshamand Company Ltd.

Shulte, A.P. (Ed.) (1981). Teaching statistics and probabilitv. Reston,

Virginia: National Council of Teachers of Mathematics.

Tanur, J. e t al (Eds.) (1978). Statistics: A guide t o t h e unknown (2nd

ed. 1. San Francisco, California: Holden-Day.