The Goal: a process of Ongoing Improvement

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 quality control in a different way.’’
I ask, "You mean we should put Q.C. in front of the bottlenecks?’’
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 inspectors?’’
"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?’’
"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 calculating 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. They 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 according to standard
accounting procedures
. . . without
considering 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 million per month.’’

---

"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 production?’’
"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 processing 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 present money for future money; the

---

question is, can your cash flow sustain it? In your case, absolutely not.’’
"He’s right,’’ admits Lou.
"Then make the bottlenecks work only on what will contribute to throughput