Overview: This exercise is best slotted before, after, or within sessions that focus on

operational performance, capacity and bottlenecks, improving a process, and process

variability. It may also be helpful in elective courses that require additional exploration
of basic concepts. In Service Operations and Supply Chain Management courses, this
exercise introduces what students often consider to be dry concepts. One criticism of
operations-related courses that are approached from a managerial or strategic
perspective is that they lack analytical rigor. However, students generally enjoy courses
in which tools and analytical techniques are used, as students often perceive them as
chances to gain an analytical edge. These slides present a simple way of illustrating the
relevance of metrics used in process analysis.

1
Major points:
Start with a description of a simplified car wash. As cars drive into the car wash,
they line up and wait to be washed. This car wash can only clean one car at a
time and on average can clean a car in three minutes. After describing the car
wash, proceed with the questions below. If these concepts have been taught
before, this can be a review of previous readings or classes.
Highlight:
While walking through each question, challenge the students to link the terms
to the car wash example.
The specific Key Concepts of Process Analysis evaluated in the exercise are:
Cycle Time = Average time between completion of successive units; expressed
as time/unit (e.g., a vehicle exits the car wash every five minutes)
Throughput Rate = Average number of units processed per time unit; expressed
as units/time (e.g., the vacuum station can process 20 cars/hour)
Throughput Time = Average time a unit spends in the system (including all
waiting times); expressed as time (e.g., every car spends 15 minutes inside the
car wash)
Capacity Utilization = Average output relative to the maximum output possible,
expressed as a percentage (e.g., the capacity utilization is 45%, accounting for
the average number of interruptions and disruptions)
Transition:
Today we will go through two versions of this car wash scenario.

2
Major points:
In this exercise, the student plays the role of a car wash operations manager.
The time frame is a regular eight-hour workday: 9:00 am to 5:00 pm. The
customer arrival rate varies according to the time of day, with a peak in demand
from 12:00 pm to 1:00 pm. As the operations manager, the student has the job
of allocating the entire budget of $3 million to the vacuum, the machine wash,
or the hand dry stations, in order to maximize efficiency and thereby improve
overall profitability. (Pg. 3)
Highlight:
No variation and the objective is to maximize profits
Transition:
Before we begin, we will discuss a few key concepts.

3
Major points:
Here we review what a bottleneck is and how it effects the throughput. In
manufacturing, a bottleneck describes what happens when the demand being
placed on a resource exceeds its capacity. Operations in which capacity is
greater than demand are considered non-bottleneck.

4
Main Point:
At this time, you can take questions and suggestions from the students on what
the bottleneck would be for the car wash situation.
Expected questions:
Most students will speculate about the location of the bottleneck and will
mention the concepts of Throughput Rate, Cycle Time, Throughput Time, and
Capacity Utilization. Quickly touching on this basic terminology lays the
groundwork for discussion. During the introduction, it is important to emphasize
assumptions (e.g., demand is steady) to steer the conversation away from the
demand drivers of profitability and concentrate on operational issues.

5
Major point:
The primary constraint is the budget.
Transition:
Students can play challenge 1 after the instructor presents this slide.
Estimated total time for play-through: 10 minutes
It costs $675 per day to run the car wash, so at the beginning of every day (run),
the total profit is $675
A wash costs $5 per car; the cost does not depend on the amount of time spent
in each station
Each car washed yields $10 of revenue
The underlying equation of profitability is: Profit = Cars Washed x ($10 $5)
$675
Optional tie-in with The Goal - Herbie
A troop of Boy Scouts all walk in a line and cannot pass each other. This is the
notion of dependent events. The boys also walk at different rates. This is
variation. The leader of the Scouts tries several strategies to keep them
together, walk quickly, and not waste energy.
The first strategy is to just walk in random order. The problem becomes that
some boys go faster than others and the line gets gaps between boys in certain
places and bunches of boys in other places.
The second strategy is to let the fastest boys go first. The problem with this
strategy is that they spread out too much and the slow boys go even slower

6
because they have less incentive to move faster.
Finally, they try putting the slowest boy at the front of the line; this boys name is
Herbie. In manufacturing language, Herbie is a bottleneck. The line of boys now
can only move as fast as Herbie goes. This strategy starts working. All the boys
give Herbie encouragement to go faster. Herbie works harder, but then he starts
to tire. The boys realize that they need to relieve Herbie of some of the things he
is carrying so that he doesnt have to work so hard. They take his backpack and
the whole line starts moving faster.

6
Major points:
After all students have completed Challenge 1, the instructor may wish to
conduct a short debriefing session and reveal the optimal solution: The
maximum profitability of Challenge 1 is $825. Or, the instructor may advise
students to advance directly to Challenge 2 and he or she will conduct a
comprehensive debriefing after both challenges are completed. Each instructor
must decide when to stop the exercise and initiate a discussion of the learning
points, with or without the aid of the aforementioned debrief PowerPoint deck.
Highlight:
The basic messages are that capacity utilization must be balanced among the
stations and that a dollar invested in a non-bottlenecked operation gets wasted.
At 37.5 cars/hour for each station, no bottleneck is created. As one car finishes
at one station, it can move into the next station immediately.

7
Main Points:
What is the difference between even investment and optimal investments?
The students should be able to see how maximizing utilization and unequal
investment led to higher profits.
Walk through how Littles Law can be used in this scenario.
Work-In-Process (WIP) = Average number of units in the system;
expressed as units (e.g., five cars are being serviced at any given time)
Analytically, Littles Law accounts for this basic relationship
among variables:
WIP = Throughput Rate x Throughput Time
Heres a simple way to explain this concept to students: If you eat one apple per day
(throughput rate) and you have 10 apples (WIP), it will take you 10 days to finish all the
apples. If the system can process 20 cars per hour (throughput rate), and each car
spends 15 minutes or 1/4 hour (throughput time) in the system, then on average five
cars are in the system at once (WIP). Of course, this assumes steady arrivals, no major
disruptions, and no variability in the processing times, as in Challenge 1.

8
Main Points:
This slide is a placeholder to show student performance. Exit the presentation
and show the results using the facilitator screens (review pages 810 of the
teaching note for more information on these screens).
Highlight:
Class averages and general trends
Ask the students, What were the strategies you employed?

9
Main Points:
This slide illustrates what the potential responses are for different investments.
Highlight:
Profit maximized at $825
Different combinations can create similar profits
Transition:
Next, we will move on to challenge number 2, which takes into account
variation in demand.

10
Major points:
Challenge 2, labeled Variation, relaxes the previous assumptions and exploits
variability in all its forms. This scenario enables students to understand the
complexity of a real-world operation and to see firsthand how a simulation is
ideal for visualizing conflicting trade-offs. In Challenge 2, students must invest
the entire $3 million budget not only to increase capacity but, just as crucially, to
reduce variability in each of the stations. In contrast to Challenge 1, which rests
on the principles of Littles Law, Challenge 2 uses the principles of variability (for
both demand and process).
Highlight:
There is now variability in the system that mimics reality
The objective is still to maximize profits
Expected questions:
Will we have to use the entire $3 million? Answer: Yes, the entire budget
must be used in this scenario.
Transition:
After briefing students on the exercise, have them try Challenge 2 on their own.

11
Main Points:
Discuss the change in variability in from Challenge 1 to Challenge 2
Highlight:
Challenge 2 explicitly evaluates how variability influences process performance
Each station now shows variability

12
Main Point:
Students have the option to increase capacity at any of the three processes, but
a cost (investment in stations) is required to accomplish the desired outcome.
The problem is somewhat complex because the stations are linkedthe output
of one station becomes the input of the next (i.e., cars must proceed from the
vacuum station to machine wash and then to hand dry). Therefore, balance
among stations and the need to invest in stations that become bottlenecks
becomes relevant.
Highlight:
Rated capacity is the theoretical maximum capacity (e.g., maximum number of
units that can be produced under perfect conditions) . Usable capacity is the
actual capacity after factoring random variables that diminish the maximum
(e.g., breakdowns, quality of inputs)

13
Main Points:
An Organizations productive capacity is affected by many different factors.
Some can be controlled by management, while others are not. These factors
create the gap between what is the rated capacity and usable capacity.
Discussion:
Discuss with students examples of what can affect utilization.
Example of controlled utilization factors:
Acquisition and supervision of land and physical resources
Utilization of labor
Examples of non-controllable utilization factors:
Personnel issues
Weather events
Political issues
Unpredicted interruptions

14
15
Main Point:
The balanced capacity solution from Challenge 1 will not necessarily generate
the maximum profitability in Challenge 2.
Highlight:
The displayed strategy is the optimal strategy for Challenge 1. Note that
balancing among the three stations does not result in optimal allocation. The
fact that Vacuum and Wash Capacity are fully utilized while Hand Dry is at times
not fully utilized is a key clue that investment in Hand Dry can be shifted to the
other stations. Main reason for this is the variability which was not the case in
Challenge 1.
Expected questions:
What is the drop-off at the end of the graph? Answer: The daily closure of the
car wash.

16
Main Point:
In Challenge 2, as investment increases, capacity increases and variability
decreases.
Highlight:
The output of one station becomes the input of the next.
Discussion:
How can investments change variability?

17
Main Points: The following graphs illustrate that different combinations of investment
can yield the same result (i.e., performance). Students might get similar results for
different investment allocations. At first it might seem confusing but the explanation is
simply that the simulation calculates revenues based on complete units (i.e., a car
wash). This chart illustrates how the system generates a Profit of $675 for the 11
scenarios (each with a different allocation to the three stations) because in each one of
the scenarios the system finished 270 cars washes. This is NOT the chart for the optimal
allocation just an example where different combinations could generate same results.
The optimal allocation results in a profit of $825 with 300 cars washed.

Highlight:
The calculations are based on discrete, not continuous units, because revenue is
based on whole units. You cant have a partial car wash.

18
Main Points:
This Chart summarizes the behavior of Variability for the two Challenges. In
Challenge 1 the initial allocations (before investments) had the vacuum at
36cars/hour, the Washer at 30 cars/hour and the Dryer at 24 cars/hour. Higher
or lower investment would increase or decrease the capacity in a linear fashion
(i.e., no variability) and that is why at optimality (after investing $1M, $0.5M and
$1.5M respectively) the capacity of all stations is balanced at 37.5 cars/hour.
For Challenge 2 that is not the case. Capacity does not increase or decrease
linearly with investment. The shape of the curve captures the behavior for
each station. Investing more (or less) in any of the stations increases (or
decreases) the resulting capacity (in car wash/unit). Then for example, more
investment might imply a better technology that makes the capacity to increase
faster. In Challenge 1 increases are linear with the investment.
Highlight:
The behavior of the capacity/investment tradeoff is not always linear. Finding
the balance is significantly more complicated in the presence of variability.
That is why the mean should not be the only important descriptor. Most
environments are evaluated using the mean as the descriptor. However, a
mean can be misleading. Even if a mean forecast is perfect, on its own the
number cannot reveal, for example, that one five-day period exceeded the
mean and another five-day period performed below the mean.
Discussion Questions:

19
Ask the students where you should invest and how would that change capacity
in light of the variability we observe in Challenge 2?
Example: Investments in new technology, training, learning curve

19
Main Points:
This slide is meant to create a discussion; answers will be given later on.
After each question, it is recommended that you garner feedback from students.

20
Main Points:
This slide is a placeholder to show student performance. Exit the presentation
and use the facilitator screens to display results (review pages 8-10 of the
teaching note for more information).

Highlight: Class averages and general trends

Ask the students, What were the strategies you employed?

21
22
Main Point:
Capacity is NOT balanced across the three stations.
In Challenge 2, because of the explicit variability and interconnection among the
three stations, there are many potential results. Notably, optimal profitability
does not derive from a unique combination of investment options; it is possible
to get the same results with different combinations of investments.
Highlight:
Littles Law is very good for a foundation, but there is implicit behavior in the
real world that is more complex.

23
Main Point:
Top right corner shows the total profit, which is lower in this challenge.
Utilization is 100% through almost the entire day.
At 1.69 min/car, cycle time is higher than challenge 1.

24
25
Main Point:
Demonstration of the optimal strategy for both challenge 1 and challenge 2
Highlight:
At the optimal strategies for Challenge 1 and Challenge 2, profits arent the
same. Ask the students what can cause these lower profits in Challenge 2.
Transition:
The next slide will review variability and its effects on profit due to wait time.

26
Main Point:
When using the same Allocation Strategy in Challenge 2 as in Challenge 1
Profitability goes from $825 to $655 (a reduction of 20%). The optimal
Allocation Strategy for Challenge 2 generates a profitability of $690 which is still
lower than the optimal you obtained in Challenge 1 but higher than the
profitability you would obtain if just applied the same allocation from the
optimal solution at Challenge 1 ($690 versus $655). The reduction in
Profitability for the optimal allocation of Challenge 2 relative to the optimal
allocation of Challenge 1 ($690 versus $825) is due to the inefficiencies created
by the variability. This is a very important learning. Variability definitely
represents a cost for every system and that is why we need to understand its
impact.
In the presence of variability, there is a trade-off between high throughput rates
and low manufacturing (i.e., car wash) lead times. This trade-off becomes more
severe as process variability increases. In the presence of variability, queue time
can become many times larger than processing time as utilization approaches
100%.

Highlight:
The average wait time is defined as the average amount of time that a car must
wait in the queue before being processed. A car that arrives to find an empty
system suffers no wait at all but those zeroes in waiting time are still used in

27
 calculating average wait time.
Waiting time increases as the utilization is closer to 100% creating queues.
Variability makes waiting time unpredictable. Sometimes it is the customer
waiting for service and sometimes the other way around (i.e., a station waiting
while the previous is finishing). This increase in wait time ultimately effects the
maximum profit a system can obtain. The comparison of resulting leadtimes for
the two challenges confirms the additional impact in the system and one of the
reasons profitability is lower for the optimal solution of Challenge 2 relative to
Challenge 1. Remember that the optimal allocation for Challenge 1 finished
300 cars while the optimal in Challenge 2 finishes 273. This is illustrated in the
next slide.
Tie-in:
Recommended tie in with the The Impact of Variability on Process
Performance, HBP Product No. 8228.

27
Main Point:
This slide is a summary of the different results for Challenge 1 and Challenge 2
where all comparisons on performance (profitability and number of cars) are
captured. As discussed before, variability across the stations eliminates a
continuous movement of cars through the system, and the resulting irregular
build-up of WIP can create different throughputs (and therefore varying
profitability). The chart only illustrates the relationship between different
allocations for the different stations from the optimal solution for Challenge 2.

Highlight:
Implication of Variability: As the variability in a process becomes more severe, it
becomes more and more difficult to achieve high utilization (i.e., higher
throughput) and low WIP inventory (i.e., cars waiting in the different stations)
simultaneously.

28
Main Points:
Responses to variability can include both external and internal policies.
Possible responses to variability include turning away business and foregoing
potential production, building extra capacity, or simply allowing queues and
inventories to be large.

29
Main Points:
The OM Triangle can be used as a framework to explain the relationship
between Capacity and Speed of Response. The curves capture the variability in
the system: Blue curve is no Variability while Purple curve high Variability.
Under the blue curve the system could maximize the capacity utilization to a
much higher number while maintaining a low level of WIP inventory. Imagine
customers arriving at a rate of 4/hour with a processing time of 5/hour and NO
variability. Then the system could operate at 100% capacity utilization and the
WIP would be very low since the system can process all customers as they
arrive. Under the Purple curve (high variability) if the arrival rate is 4/hour and
the processing is 5/hour BUT with a lot of variability at either arrival or
processing times, there are going to be queues and the service would be
delayed since the system would be fast in some cases and very slow in others.
That is why the only way you could equate the service of the Blue curve and
the Purple curve is by increasing the capacity of the system (e.g., to 10/hour) so
the variability is absorbed and that results in a lower capacity utilization. The
third point in the triangle, High Inventory is the result when you try to
maximize the capacity utilization (with high variability) then the system will
accumulate a higher WIP and therefore a slower speed of response. That is why
we argue that capacity, inventory and variability have an explicit trade-off.
Reducing Variability is equivalent to reducing investment in Capacity and
reduction in WIP inventory. Variability is always a cost to the system. This is

30
 explain in a lot of detail in the reading The Curse of Variability.

Highlight:
The red dot with the label Low Variability represents Challenge 1,
The red dot with the label Extra Capacity represents Challenge 2. With more
variability and a similar system the resulting output is lower (less cars processed)
and therefore lower profitability.

30
31