The Goal: A Process of Ongoing Improvement, Third Revised Edition
Page 22
in a moment. But one more thing," says Jonah. "I want to know
where you do quality inspection on bottleneck parts."
I explain to him that most inspection is done prior to final
assembly.
"Show me," says Jonah.
So we go to an area where we do quality inspections. Jonah
asks about bottleneck parts that we reject. Immediately, Bob
points to a pallet stacked with shiny steel parts. On top of them is
a pink sheet of paper, which indicates rejection by Quality Con-
trol, or Q.C. as it's known. Bob picks up the job jacket and reads
the forms inside.
"I'm not sure what's wrong with these, but they must be
defective for some reason," says Bob.
Jonah asks, "Did these parts come through a bottleneck?"
"Yeah, they did," says Bob.
E.M. Goldratt
The Goal: A Process of Ongoing Improvement
Captured by Plamen T.
163
"Do you realize what the rejection by Q.C. has done to you?"
asks Jonah.
"It means we have to scrap about a hundred parts," says
Bob.
"No, think again," says Jonah. "These are bottleneck parts."
It dawns on me what he's getting at.
"We lost the time on the bottleneck," I say.
Jonah whirls toward me.
"Exactly right!" he says. "And what does lost time on a bottleneck mean? It means you have lost throughput."
"But you're not saying we should ignore quality, are you?"
asks Bob.
"Absolutely not. You can't make money for long without a
quality product," says Jonah. "But I am suggesting you use qual-
ity control in a different way."
I ask, "You mean we should put Q.C. in front of the bottle-
necks?"
Jonah raises a finger and says, "Very perceptive of you. Make
sure the bottleneck works only on good parts by weeding out the
ones that are defective. If you scrap a part before it reaches the
bottleneck, all you have lost is a scrapped part. But if you scrap
the part after it's passed the bottleneck, you have lost time that
cannot be recovered."
"Suppose we get sub-standard quality downstream from the
bottleneck?" says Stacey.
"That's another aspect of the same idea," says Jonah. "Be
sure the process controls on bottleneck parts are very good, so
these parts don't become defective in later processing. Are you
with me?"
Bob says, "Just one question: where do we get the inspec-
tors?"
"What's wrong with shifting the ones you already have to the
bottlenecks?" asks Jonah.
"That's something we can think about," I tell him.
"Good. Let's go back to the offices," says Jonah.
We go back to the office building and meet in the conference
room.
"I want to be absolutely sure you understand the importance
of the bottlenecks," says Jonah. "Every time a bottleneck finishes a part, you are making it possible to ship a finished product. And
how much does that mean to you in sales?"
E.M. Goldratt
The Goal: A Process of Ongoing Improvement
Captured by Plamen T.
164
"It averages around a thousand dollars a unit," says Lou.
"And you're worried about spending a dollar or two at the
bottlenecks to make them more productive?" he asks. "First of all, what do you think the cost of, let's say, the X machine is for one
hour?"
Lou says, "That's well established. It costs us $32.50 per
hour."
"And heat-treat?"
"That's $21 per hour," says Lou.
"Both of those amounts are incorrect," says Jonah.
"But our cost data—"
"The numbers are wrong, not because you have made a cal-
culating error, but because the costs were determined as if these
work centers existed in isolation," says Jonah. "Let me explain: when I was a physicist, people would come to me from time to
time with problems in mathematics they couldn't solve. Thev
wanted me to check their numbers for them. But after a while I
learned not to waste my time checking the numbers—because the
numbers were almost always right. However, if I checked the
assumptions, they were almost always wrong."
Jonah pulls a cigar out of his pocket and lights it with a
match.
"That's what's going on here," he says between puffs. "You
have calculated the cost of operating these two works centers ac-
cording to standard accounting procedures . . . without consid-
ering the fact that both are bottlenecks."
"How does that change their costs?" asks Lou.
"What you have learned is that the capacity of the plant is
equal to the capacity of its bottlenecks," says Jonah. "Whatever the bottlenecks produce in an hour is the equivalent of what the
plant produces in an hour. So ... an hour lost at a bottleneck is
an hour lost for the entire system."
"Right, we're with you," says Lou.
"Then how much would it cost for this entire plant to be idle
for one hour?" asks Jonah.
"I really can't say, but it would be very expensive," admits
Lou.
"Tell me something," asks Jonah. "How much does it cost
you to operate your plant each month?"
Lou says, "Our total operating expense is around $1.6 mil-
lion per month."
E.M. Goldratt
The Goal: A Process of Ongoing Improvement
Captured by Plamen T.
165
"And let's just take the X machine as an example," he says.
"How many hours a month did you say it's available for produc-
tion?"
"About 585," says Ralph.
"The actual cost of a bottleneck is the total expense of the
system divided by the number of hours the bottleneck produces,"
says Jonah. "What does this make it?"
Lou takes out his calculator from his coat pocket and
punches in the numbers.
"That's $2,735," says Lou. "Now wait a minute. Is that
right?"
"Yes, it's right," says Jonah. "If your bottlenecks are not working, you haven't just lost $32 or $21. The true cost is the cost
of an hour of the entire system. And that's twenty seven hundred
dollars."
Lou is flabbergasted.
"That puts a different perspective on it," says Stacey.
"Of course it does," says Jonah. "And with that in mind, how do we optimize the use of the bottlenecks? There are two principal themes on which you need to concentrate . . .
"First, make sure the bottlenecks' time is not wasted," he
says. "How is the time of a bottleneck wasted? One way is for it to
be sitting idle during a lunch break. Another is for it to be pro-
cessing parts which are already defective—or which will become
defective through a careless worker or poor process control. A
third way to waste a bottleneck's time is to make it work on parts
you don't need."
"You mean spare parts?" asks Bob.
"I mean anything that isn't within the current demand," he
says. "Because what happens when you build inventory now that
you won't sell for months in the future? You are sacrificing pres-
ent money for future money; the question is, can your cash flow
&nb
sp; sustain it? In your case, absolutely not."
"He's right," admits Lou.
"Then make the bottlenecks work only on what will contrib-
ute to throughput today . . . not nine months from now," says Jonah. "That's one way to increase the capacity of the bottlenecks. The other way you increase bottleneck capacity is to take
some of the load off the bottlenecks and give it to non-bottle-
necks."
I ask, "Yeah, but how do we do that?"
E.M. Goldratt
The Goal: A Process of Ongoing Improvement
Captured by Plamen T.
166
"That's why I was asking those questions when we were out
in the plant," he says. "Do all of the parts have to be processed by the bottleneck? If not, the ones which don't can be shifted to non-bottlenecks for processing. And the result is you gain capacity on
your bottleneck. A second question: do you have other machines
to do the same process? If you have the machines, or if you have a
vendor with the right equipment, you can offload from the bottle-
neck. And, again, you gain capacity which enables you to increase
throughput."
I come into the kitchen for breakfast the next morning and
sit down to a big steaming bowl of my mother's oatmeal . . .
which I have hated ever since I was a kid. I'm staring at the
oatmeal (and the oatmeal is staring back) when Mom/Grandma
asks, "So how did everything go last night?"
I say, "Well, actually, you and the kids were on the right
track at dinner."
"We were?" asks Dave.
"We need to make the Herbies go faster," I say. "And last
night Jonah pointed out some ways to do that. So we learned a
lot."
"Well, now, isn't that good news," says my mother.
She pours a cup of coffee for herself and sits down at the
table. It's quiet for a moment. Then I notice that Mom and the
kids are eyeing each other.
"Something wrong?" I ask.
"Their mother called again last night while you were gone,"
says my mother.
Julie has been calling the kids regularly since she left. But for
whatever reason of her own, she still won't tell them where she is.
I'm debating whether to hire a private detective to find out where
she's hiding.
"Sharon says she heard something when she was on the
phone talking," says my mother.
I look at Sharon.
"You know that music Grandpa always listens to?" she says.
I say, "You mean Grandpa Barnett?"
"Uh-huh, you know," she says, "the music that puts you to
sleep, with the—what are they called?"
"Violins," says Dave.
E.M. Goldratt
The Goal: A Process of Ongoing Improvement
Captured by Plamen T.
167
"Right, the violins," says Sharon. "Well, when Mom wasn't
talking, I heard that on the phone last night."
"I heard 'em too," says Dave.
"Really?" I say. "That's very interesting. Thank you both for noticing that. Maybe I'll give Grandma and Grandpa Barnett another call today."
I finish my coffee and stand up.
"Alex, you haven't even touched your oatmeal," says Mom.
I lean down and kiss her on the cheek. "Sorry, I'm late for
school."
I wave to the kids and hurry to grab my briefcase.
"Well, I'll just have to save it so you can eat it tomorrow,"
says my mother.
E.M. Goldratt
The Goal: A Process of Ongoing Improvement
Captured by Plamen T.
168
20
Driving to the plant, I pass the motel where Jonah stayed last
night. I know he's long gone—he had a 6:30 A.M. flight to catch. I
offered to pick him up this morning and drive him to the airport,
but (lucky for me) he refused and said he'd take a cab.
As soon as I get to the office, I tell Fran to set up a meeting
with the staff. Meanwhile, I start to write down a list of the actions
Jonah suggested last night. But Julie comes to mind and won't
leave. I close my office door and sit down at my desk. I find the
number for Julie's parents and dial it.
The first day after Julie left, her parents called to ask me if I
had heard anything. They haven't called back since. A day or two
ago, I tried getting in touch with them to find out if they had
heard anything. I called in the afternoon and I talked to Julie's
mother, Ada. She said she didn't know where Julie was. Even
then, I didn't quite believe her.
Now Ada answers again.
"Hi, this is Alex," I tell her. "Let me talk to Julie."
Ada is flustered. "Well, um, ah ... she isn't here."
"Yes, she is."
I hear Ada sigh.
"She is there, isn't she," I say.
Finally Ada says, "She does not want to talk to you."
"How long, Ada? How long has she been there? Were you
lying to me even that Sunday night when I called?"
"No, we were not lying to you," she says indignantly. "We had no idea where she was. She was with her friend, Jane, for a
few days."
"Sure, and what about the other day when I called?"
"Julie simply asked me not to say where she was," says Ada,
"and I shouldn't even be telling you now. She wants to be by
herself for a while."
"Ada, I need to speak with her," I say.
"She will not come to the phone," says Ada.
"How do you know until you've asked?"
The phone on Ada's end is put down on the table. Footsteps
fade away and return a minute later.
E.M. Goldratt
The Goal: A Process of Ongoing Improvement
Captured by Plamen T.
169
"She says she'll call you when she's ready," says Ada.
"What does that mean?"
"If you hadn't neglected her all these years, you wouldn't be
in this situation," she says.
"Ada—"
"Good-bye," she says.
She hangs up the phone. I try calling back right away, but
there is no answer. After a few minutes, I force my mind back to
getting ready to talk to the staff.
At ten o'clock, the meeting starts in my office.
"I'd like to know what you think about what you heard last
night," I say. "Lou, what was your reaction?"
Lou says, "Well . . . I just couldn't believe what he was say-
ing about an hour of a bottleneck. I went home last night and
thought it over to see if it all made sense. And, actually, we were
wrong about a lost hour of a bottleneck costing $2,700."
"We were?" I ask.
"Only eighty percent of our products flow through the bot-
tlenecks," says Lou as he takes a piece of note paper from his
shirt pocket. "So the truer cost ought to be eighty percent of our
operating expense, and that comes to $2,188 an hour—not
$2,735."
"Oh," I say. "I suppose you're right."
Then Lou smiles.
"Nevertheless," he says, "I have to admit it was quite an eye-opener to look at the situation from that perspective."
"I agree," I say. "What about the rest of you?"
I go from person to person around the office asking for reac-
tions, and we're all pretty much in agreement. Even so, Bob
seems hesitant about co
mmitting to some of the changes Jonah
was talking about. And Ralph isn't sure yet where he fits in. But
Stacey is a strong advocate.
She sums up, saying, "I think it makes enough sense to risk
the changes."
"Although I'm nervous about anything that increases operat-
ing expense at this point in time," says Lou, "I agree with Stacey.
As Jonah said, we may face a bigger risk just staying on the path
we've been following."
Bob raises one of his meaty hands in preparation for a com-
ment.
E.M. Goldratt
The Goal: A Process of Ongoing Improvement
Captured by Plamen T.
170
"Okay, but some of what Jonah talked about will be easier
and faster to make happen than the rest," he says. "Why don't we go ahead with the easier things right away and see what kind of
effect they have while we're developing the others."
I tell him, "That sounds reasonable. What would you do
first?"
"I think I'd wanna move the Q.C. inspection points first, to
check parts going into the bottlenecks," says Bob. "The other
Q.C. measures will take a little time, but we can have an inspector
checking pre-bottleneck parts in no time—by the end of today if
you want."
I nod. "Good. What about new rules for lunch breaks?"
"We might have a squawk or two from the union," he says.
I shake my head. "I think they'll go along with it. Work out
the details and I'll talk to O'Donnell."
Bob makes a note on the paper pad on his lap. I stand up
and step around the desk to emphasize what I'm about to say.
"One of the questions Jonah raised last night really struck
home for me," I tell them. "Why are we making the bottlenecks
work on inventory that won't increase throughput?"
Bob looks at Stacey, and she looks back at him.
"That's a good question," she says.
Bob says, "We made the decision—"
"I know the decision," I say. "Build inventory to maintain
efficiencies." But our problem is not efficiencies. Our problem is
our backlog of overdue orders. And it's very visible to our cus-
tomers and to division management. We positively must do some-
thing to improve our due-date performance, and Jonah has given
us the insight on what that something has to be.
"Until now, we've expedited orders on the basis of who's
screamed the loudest," I say. "From now on, late orders should
get first priority over the others. An order that's two weeks late
gets priority over an order that's one week late, and so on."