Question #5: If the restaurant runs a sale and the customer arrival rate increases by 20%, how would this change the total time expected to serve a customer? How would this change the average number of cars in the drive thru-line?
Transcribed Image Text: Analytics Exercise: Processing Customer
Orders
Analyzing a Taco Bell Restaurant
The following scenario was written by a reporter who became
a Taco Bell worker for a few hours to experience what it's
like to work at one of the most high-tech quick-serve res-
taurant chains in the world. As you read, visualize how you
could analyze a Taco Bell using the queuing models that we
discussed in this chapter. After the scenario, we will give you
some hints related to how you can model the Quick Service
(QS) restaurant and then ask a series of questions related to
your model.
It must always be, "Hi, how are you today?" Never,
"Hi, how are you?" "Hi, how's it going?" or "Wel-
come to Taco Bell." Never, "What will it be today?"
or, even worse, "What do you want?" Every Taco Bell
Service Champion memorizes the order script before his
first shift. The folks who work the drive-thru windows at
the Taco Bell here in Tustin, California, about 35 miles
south of Los Angeles, and everywhere else, are called
Service Champions. Those who work the food produc-
tion line are called Food Champions.
You think you know it-"Hi, how are you today?"
It seems easy enough. And you follow that with, "You
can order when you're ready," never "Can I take your
order?" The latter puts pressure on the driver, who might
be a distracted teenager busy texting her friend or a soc-
cer mom with a half-dozen kids in the van. "They don't
need the additional pressure of a disembodied voice
demanding to know their order," explains Mike Harkins.
Harkins, 49, is vice-president of One System Operations
for Taco Bell, which means he spends all day, every day,
thinking about the kitchen and the drive-thru.
He has been prepping me for my debut at the window.
Getting ready, I wash my hands, scrubbing for the man-
dated 20 seconds; slide on rubber gloves; and don the
three-channel headset that connects me to the ordering
station out in the lot, as well as to my fellow Champions.
I take my place at the window. I hear the ding indicating
a customer has pulled into the loop around the restau-
rant, and I immediately ask, "Hi, how's it going?"
It gets worse from there. As a Service Champion, my
job is to say my lines, input the order into the proprietary
point of sale (POS) system, prepare and make drinks like
Limeade Sparklers and Frutista Freezes, collect bills or
credit cards, and make change. I input Beefy Crunch
Burritos, Volcano Burritos, Chalupas, and Gorditas. My
biggest worry is that someone will order a Crunchwrap
Supreme, a fast-food marvel made up of two kinds of
tortillas, beef, cheese, lettuce, tomatoes, and sauces,
all scooped, folded, and assembled into a handheld,
multiple-food-group package, which then gets grilled
for 27 seconds. This actually doubles the time it takes
to prepare a normal order. An order for a Crunchwrap
Supreme, the most complex item on the menu, sometimes
requires the Service Champion to take up position on the
Transcribed Image Text: food production line to complete it in anything like the
164 seconds that Taco Bell averages for each customer,
from driving up to the ordering station to pulling away
from the pick-up window.
Above me on the wall, a flat-screen display shows
the average time of the last five cars at either the order
station or the pick-up window, depending on which is
slowest. If the number is red, as it is now, that means
one, or both, of the waits is exceeding 50 seconds, the
target during peak periods. It now shows 53 seconds,
on its way to 60, 70... and then I stop looking. The
high-pitched ding that announces each new customer
becomes steady, unrelenting, and dispiriting-85 cars
will roll through over the peak lunch rush. And I keep
blowing the order script.
I fall behind so quickly and completely that restau-
rant manager Amanda Mihal, a veteran of 12 years in
the QSR business (Quick Serve Restaurant, the acronym
for an industry that makes acronyms for everything), has
to step in. "You'll get it," Amanda says as she fixes an
order that I have managed to screw up. "Eventually,"
Every Taco Bell has two food production lines, one
dedicated to the drive-thru and the other to servicing
liner in
the walk-up counter. Working those lines is no easier
than wearing the headset. The back of the restaurant
has been engineered so that the Steamers, Stuffers, and
Expeditors, the names given to the Food Champions
who work the pans, take as few footsteps as possible
during a shift. There are three prep areas: the hot hold-
ing area, the cold holding area, and the wrapping ex-
pediting area. The Stuffer in the hot holding area stuffs
the meat into the tortillas, ladling beef with Taco Bell's
proprietary tool, the BPT, or beef portioning tool. The
steps for scooping the beef have been broken down into
another acronym, SST, for stir, scoop, and tap. Flour
tortillas must be cooked on one side for 15 seconds and
the other for five.
When I take my place on the line and start to prepare
burritos, tacos, and chalupas-they won't let me near a
Crunchwrap Supreme-it is immediately clear that this
has been engineered to make the process as simple as
possible. The real challenge is the wrapping. Taco Bell
once had 13 different wrappers for its products. That has
been cut to six by labeling the corners of each wrapper
differently. The paper, designed to slide off a stack in
single sheets, has to be angled with the name of the item
being made at the upper corner. The tortilla is placed
in the middle of the paper and the item assembled from
there until you fold the whole thing up in the wrapping
expediting area next to the grill. "We had so many wrap-
pers before, half a dozen stickers; it was all costing us
seconds," says Harkins. In repeated attempts, I never get
the proper item name into the proper place. And my bur-
ritos just do not hold together.
With me on the line are Carmen Franco, 60, and
Ricardo Alvarez, 36. The best Food Champions can
prepare about 100 burritos, tacos, chalupas, and gordi-
tas in less than half an hour, and they have the 78-item
menu memorized. Franco and Alvarez are a precise and
frighteningly fast team. Ten orders at a time are dis-
played on a screen above the line, five drive-thrus and
five walk-ins. Franco is a blur of motion as she slips
out wrapping paper and tortillas, stirs, scoops, and taps,
then slides the items down the line while looking up at
the screen. The top Food Champions have an ability to
scan through the next five orders and identify those that
require more preparation steps, such as Grilled Stuffed
Burritos and Crunchwrap Supremes, and set those up
before returning to simpler tacos and burritos. When
Alvarez is bogged down, Franco slips around him, and
slides Crunchwrap Supremes into their boxes.
At the drive-thru window in Tustin, I would have
shaken off the headset many orders ago had it not been
for manager Mihal's support, but I'm hanging in there.
After a while, I do begin to detect a pleasing, steady
rhythm to the system, the transaction, the delivery of the
food. Each is a discrete, predictable, scripted interaction.
When the order is input correctly, the customer drives up
to the window, the money is paid, the Frutista Freeze or
Atomic Bacon Bombers (a test item specific to this Taco
Bell) handed over, and you send people on their way with
a smile and a "Thank you for coming to Taco Bell," you
feel a moment of accomplishment. And so does Harkins,
for it has all gone exactly as he has planned.
Then a ding in my headset.
"Um, hello?"
Idiot, I think to myself, I've blown the script again.
Source: Karl Taro Greenfeld, Bloomberg Business Week Features Section,
May 5, 2011
Modeling the Restaurant
In the scenario, they indicate that it takes about 164 seconds
on average to serve a customer during the busy lunch hour
period. Put yourself in the seat of your car getting food at
the FS restaurant. Let's assume you are using the drive-thru
window and that you will pick up the food and take it home
to eat with some friends.
You drive into the restaurant lot and notice there is a line
of cars that has formed at the order kiosk. You wait for your
turn to talk to the Customer Service Champion so that you can
place your order. The menu is sitting there in clear view, so you
can see exactly what you want. Soon it is your turn and you
quickly place your order, learn what the bill will be, and move
your car to the line at the drive-thru window. While waiting,
you get your money out and count out the exact change you
will need. After a short time, it's your turn at the window and
you give the Service Champion your money, take your drink
and food, and carefully drive out of the parking lot.
242
section 2
.
MANUFACTURING AND SERVICE PROCESSES
Think about what happened at the restaurant. First, you
waited in two lines. The first was at the order kiosk and the
second at the drive-thru window. Next, consider the work
that the restaurant needed to complete to process your order.
The Service Champion took your order and entered it in the
POS system, prepared your drink, and then when the food
was ready, collected your money, and delivered your drink
and food. One of the Food Champions prepared your food
using information from a screen that shows orders as they are
entered by the Service Champion.
The total time that it takes between when you arrive at
the restaurant until you leave is made up of the following
elements:
1 The service time for the Service Champion to process
your order
2
The service time for the Food Champion to prepare
your order
3 The waiting while the Service Champion and Food
Champion served other customers
To model this using the queuing models in the chapter assume
that you have two totally independent service processes. The
first process is the Service Champion and the second is the
Food Champion. Each process has potentially a different
mean service time per customer. The Service Champion must
serve each customer and they arrive at a particular rate. The
Food Champion prepares the individual items on the order
such as a burrito, taco, chalupa, or gorditas taco. As the orders
are taken, each individual item appears on a monitor telling
the Food Champion what should be made next. The average
time for a customer to run through the system is the sum
of the average service times (time to take the order by the
Service Champion and time to make the order by the Food
Champion) plus the sum of the expected waiting times for
the two processes. This assumes that these processes operate
totally independent of each other, which might not be exactly
true. But we leave that to a later discussion.
Assume that the queues in front of each process are large,
meaning that there is plenty of room for cars in the line be-
fore and after the order kiosk. Also, assume there is a single
Super Quiz
1 Service systems can be generally categorized ac-
cording to this characteristic that relates to the
customer.
2 A framework that relates to the customer service sys-
Service Champion and two Food Champions each operat-
ing independently and working just on the drive-thru orders.
Also, assume that the arrival pattern is Poisson, customers
are handled first-come-first-served, and the service pattern is
exponential.
Given this, answer the following questions:
1 Draw a diagram of the process using the format in
Exhibit 7.3.
2 Consider a base case where a customer arrives every
40 seconds and the Customer Service Champion can
handle 120 customers per hour. There are two Food
Champions, each capable of handling 100 orders per
hour. How long should it take to be served by the
restaurant (from the time a customer enters the kiosk
queue until her food is delivered)? Use queuing mod-
els to estimate this.
3 On average, how busy are the Customer Service
Champions and the two Food Champions?
4 On average, how many cars do you expect to have
in the drive-thru line? (Include those waiting to place
orders and those waiting for food.)
5 If the restaurant runs a sale and the customer arrival
rate increases by 20%, how would this change the
total time expected to serve a customer? How would
this change the average number of cars in the drive-
thru line?
6 Currently, relatively few customers (less than 14%)
order the Crunchwrap Supreme. What would happen
if the restaurant ran the sale, demand jumped on the
Crunchwrap Supreme, and 30% of the orders were
for this item? Take a quantitative approach to answer-
ing this question. Assume that the Customer Service
Champion never helps the Food Champions and that
these two processes remain independent.
7 For the type of analysis done in this case, what are the
key assumptions? What would be the impact on our
analysis if these assumptions were not true?
8 Could this type of analysis be used for other service-
type businesses? Give examples to support your answer.
6 Consider two queuing systems identical except for
the service time distribution. In the first system the
service time is random and distributed according to
a Poisson distribution. The service time is constant