Beer Game simulation

[Niall Palfreyman]

Hi,

this is a rather urgent question. Im planning on running the Beer Game
with a group of about 30-35 students. However I have only ONE Beer Game
board, and time is also rather short for the session. So I thought Id
cut out a lot of the time-consuming mechanics of the game by running it
as a computer simulation. Ive built a model over the weekend, but Im
not altogether happy with it. Does anyone have a ready-made Stella Beer
Game simulation which I can use? I assume there must be some out there,
but I cant find them.

Thanks,
Niall Palfreyman.
From: Niall Palfreyman <niall.palfreyman@fh-weihenstephan.de>

[zenabraham@aol.com]

Hello,

I suggest you use the "Near Beer Game," on the Forio Business Simulations
website. Its at <A HREF="http://broadcast.forio.com/sims/nearbeer">
http://broadcast.forio.com/sims/nearbeer</A>

To use the Forio description of the Near Beer game...

"This is a draft version of the Near Beer game. A cousin of the beer game.
The original Beer Game was developed to introduce students, managers and
executives to concepts of system dynamics. The game illustrates how
"structure produces behavior." Each player participates as a member of a team
that must meet its customers demands. The object of the game is to minimize
the total cost for your team. The game begins in equilibrium. Customers are
ordering 10 cases of beer and you have 10 case of inventory each week. In
week 3, your customers increase their orders to 15 cases of beer. Your
objective is to return the system to equilibrium, so that 15 cases of beer
are available in finished goods inventory week after week. You have 50 weeks
to accomplish this objective."

Good luck

Zennie Abraham
From: zenabraham@aol.com

[Richard Karash]

Niall --

I dont have such a simulation, but I know there are a couple around.
I actually experimented with this a bit using small groups several
years ago. I didnt like the results... I did not think the computer
simulation was as engaging as the actual board-play beer game.

About timing for the beer game: Yes, it takes quite a chunk of time
and its unwise to squeeze it. Assuming prior physical setup, I plan
on:
- 1:45 for instructions, practice rounds, and actual play of the
game for about 30 weeks
- 0:15 for participants to total their scores, draw graphs
- 0:45 for staff to clean up the scores and graphs while
participants have lunch or take a break
- 1:00 for debrief discussion

This totals 3 hours and 45 minutes.

You could make more boards with a magic marker, but you cant make
time as easily.

Best Regards,

-=- Rick
From: Richard Karash <Richard@Karash.com>

["Forrester, Nathan B."]

Niall,

You can download a beer game simulator from John Stermans MIT
site. See the link below. The download is free and includes
all software required to run on a PC-compatible computer
(Sorry, no Mac version.)

http://web.mit.edu/jsterman/www/SDG/MFS/simplebeer.html

My son, Matthew, developed the simulator a couple of years ago
for A.T. Kearney using Vensims Venapp Distribution Kit.

Matthew has just joined the Strategic Planning Group at
BellSouth International in Atlanta. mattforrester@uplink.net

____________________________
Nathan B. Forrester
Chief Economist
Georgia-Pacific Corporation
133 Peachtree Road
Atlanta, Georgia 30303
nbforres@gapac.com
_____________________________

["Pieter van der Hijden"]

Good Morning Niall,

When you download an evaluation copy of IThink (the business equivalent
of Stella), you will find some interesting sample models included. One
of them is the Beer Game. See the site of High Performance Systems
http://www.hps-inc.com.

Best Regards,
Pieter van der Hijden

Pieter van der Hijden - pvdh@sofos.nl - Sofos Consultancy - www.sofos.nl
- P. O. Box 94874, 1090 GW Amsterdam,

["Michael Bean"]

Niall,

Usually, I believe that its a bad idea to run the beer game as a computer
simulation.

There have been dozens of attempts at converting the beer game into a computer
simulation. In fact, Ive tried this a couple times myself. Few of these
computer versions are very much fun. Thats because the beer game was designed
to be played on a game board with other people in the room. The game structure
is simple and, because there is no computer simulating complex results in the
background, its successful at explaining how industry structures create
behavior. Making it into a computer game is like playing Jeopardy! or Monopoly
on the computer. Its more efficient, but dull. Its possible to teach supply
chain dynamics in a fun online computer simulation, but you have to build the
game in a way thats appropriate for the medium, not just recreate the beer game
on the computer.
Its almost always a bad idea to change the medium for any game or simulation.
There have been some exceptions. For example, many people enjoy on-line chess
and bridge. But this is because a player can always find a willing human
competitor. Most players would prefer to play face-to-face, but they are willing
to compromise in order to play a game of chess at 6 AM on Sunday morning before
their kids wake up.

One of Chris Crawfords five precepts in his excellent book, The Art of Computer
Game Design, is "Dont Transplant". Quoting Crawford:

"One of the most disgusting denizens of computer gamedom is the transplanted
game. This is a game design originally developed on another medium that some
misguided soul has seen fit to reincarnate on a computer....[truncated]

That being said, Ive tried (once again!) to create a fun online version of the
beer game, called the "Near Beer Game" that teaches similar lessons as the beer
game. Its designed to match the medium. You can judge how enjoyable the sim is
for yourself at:

http://broadcast.forio.com/sims/nearbeer

Ive included a copy of the underlying model for the simulation on the website,
for anyone who cares to use the model (its very simple-- only about 20
equations).

Michael Bean
Forio Business Simulation
mbean@forio.com

[Niall Palfreyman]

Hi,

Many thanks for the many suggestions and advice you all sent. Several
people sent cautions about running simulations, since they may miss the
"hands on" experience, so at the moment Im planning the following:

Ive built a 5-stage model in Stella, based on the 1-stage model
provided by Steve Peterson of HPS. The commodity is mineral water, which
Bavarians drink a lot of. (My God, Im a physicist: I never thought Id
ever use words like "commodity"!) The mineral water comes from a
manufacturer, goes to a bottler, then a distributor, then a warehouse,
then the local branch of MiniMal (German supermarket chain), then
finally down Nialls throat.

I only have 1.5 hours available, so I dont think the manual game is
feasible - Id rather save time for debriefing. Instead Ill hold a
single live game for all participants, but speeded up by using a model.
All the scoring is done automatically within the Stella model to save
time, and the history is of course graphed automatically. The students
are already grouped into 5 small groups (5-6 members), and Im planning
that each group will play a role in one communal game (one group the
manufacturer, one the bottler, and so on). I will be the communication
channel between the groups, to avoid explicit communication between the
groups, and so will guide the progress of the game. I think Ill
probably pass the information around on post-it stickers, and enter the
groups orders into the model to move from week to week.

If anyone has any comments on this setup (particularly how to facilitate
the passing of orders around the room!), Id be very glad to hear them.

Thanks very much again,
Niall Palfreyman.
From: Niall Palfreyman <niall.palfreyman@fh-weihenstephan.de>

["Bob Cavana"]

hi Niall

we have a version of the Beer Game called Mainland
beer distribution model as Case 4 (pp181- 208) in KE
Maani and RY Cavana (2000), "Systems thinking and
modelling: understanding change and uncertainty",
Pearson Education NZ (Prentice Hall), Auckland. (ISBN
1 877258 00 An ithink (Stella) version of the model is
on the accompanying CD-Rom.

this is a simplified computer version of the Beer Game,
consisting of a retail store and brewery, based on the
model developed by Prof Rolf Clark, USA.

i use this model (with a user friendly interface!) after we
have played the MIT board version of the beer game. the
model exhibits the usual dynamic patterns of behaviour
obtained from playing the manual simulation game, and
we use the computer model form to demonstrate policy
design and testing the robustness of policies.

the book (with CD-rom) is available from:

Pegasus Communications
http://www.pegasuscom.com/books.html

and Pearson education outlets.

i hope you find the case useful

regards, Bob
Dr Bob Cavana
Senior Lecturer in Decision Sciences
School of Business and Public Management, Victoria University of Wellington,
PO Box 600, Wellington, NEW ZEALAND
E-mail: bob.cavana@vuw.ac.nz
http://www.vuw.ac.nz/mgmt/staff/cavana_b/

[Niall Palfreyman]

Michael Bean:
> Usually, I believe that its a bad idea to run the beer game as a computer
> simulation...
> Its almost always a bad idea to change the medium for any game or simulation.

Yes, Im concerned about that. However this really will be a "hands on"
game, with the students making their own (group) decisions. Ill just be
the game master, with the Stella model doing the scoring automatically
in the background.

John Rodat:
> Based on what Ive seen, 1.5 hours is pretty short to run the exercise.

Yeah, I agree. Ill make the students aware beforehand that it might run
on for up to 2 hours, and as I say I hope to save time through using
Stella to automatically tally the scores.

Jim Thompson:
> If you survey the participants after the event, could you share the
> results with this listserv? It would be interesting to learn more
> about their common background and get their reactions to the experience.
Paul Newton:
> Would you share your model with me when youre done?

Will do. Ill report back next week after Ive done the run, and tell
you how the whole thing went. Anyone who wants the model is very
welcome.

John Eggers:
> Regarding communication, perhaps a designated intra-team and inter-team
> sender
eceiver would be appropriate.

Great idea - thanks John. Ill give some thought to how that might work
best.

Thanks again for all your help,
Niall Palfreyman.
From: Niall Palfreyman <niall.palfreyman@fh-weihenstephan.de>

["Keith Linard"]

Hi,

pity you are not using Powersim Constructor. Powersim allows for multi-user
simulation games and a multi-user version of the beer game is included with
the software. I have adapted this (mainly in the user interface) and used
it with undergraduate, post-graduate and senior government & defence
executives for the past 5 years (about 20 instances of the game, typically
with 4 teams of 4-6 players (some players doubling up at distributor or
factory).

The multi-user format (different teams playing retailer, wholesaler,
distributor and factory), still allows for players to gain the sense of
powerlessness induced by systemic structural issues.

Prior to using the computer simulation version, I had been utilising the
board game for several years. A key advantage, apart from time savings,
of the computer simulation over the board-game is that people were not
getting frustrated from trying to figure out the form-filling aspects.
Also, the fact that I, as game controller, had views into each of the
players screens let me follow what was happening, and provided excellent
comparative feedback for the post-game analysis.

Keith Linard
Director
UNSW Centre for Business Dynamics & Knowledge Management
University of New South Wales (ADFA Campus)
CAMPBELL ACT 2601 AUSTRALIA
Email: k-linard@adfa.edu.au