"Yes..."
"So, will anybody tell me what the process is that we should follow? What is the process mentioned in our 'process of on-go- ing improvement'? Do you think that launching several improve- ment projects is a process? We haven't done that, we have fol- lowed a process. That's what we have done."
"He's right," says Ralph in his quiet voice.
I stand up and shake Bob's hand. Everybody is smiling at him.
Then Lou asks, "What process have we followed?"
Bob doesn't hurry to answer. At last he says, "I don't know, but we definitely followed a process."
To save embarrassment I hurriedly say, "Let's find it. If we followed it, it shouldn't be too difficult to find. Let's think, what is the first thing we did?"
Before anybody has a chance to answer Ralph says, "You know, these two things are connected."
"What things?"
305
"In the 'cost world' as Alex called it, we are concerned pri- marily with cost. Cost is drained everywhere, everything cost us money. We had viewed our complex organization as if it were composed out of many links and each link is important to con- trol."
"Will you please get to the point?" Bob asks impatiently.
"Let him talk," Stacey is no less impatient.
Ralph ignores them both and calmly continues, "It's like measuring a chain according to its weight. Every link is impor- tant. Of course, if the links are very different from each other then we use the principle of the twenty-eighty rule. Twenty per- cent of the variables are responsible for eighty percent of the result. The mere fact that we all know the Pareto principle shows us to what extent Lou is right, the extent to which we all were in the cost world."
Stacey puts her hand on Bob's to prevent him from interfer- ing. ^
"We recognize that the scale has to be changed," Ralph con- tinues. "We choose throughput as the most important measure- ment. Where do we achieve throughput? At each link? No. Only at the end of all operations. You see, Bob, deciding that throughput is number one is like changing from considering weight to considering strength."
"I don't see a thing," is Bob's response.
Ralph doesn't let go, "What determines the strength of a chain?" he asks Bob.
"The weakest link, wise guy."
"So if you want to improve the strength of the chain, what must your first step be?"
"To find the weakest link. To identify the bottleneck!" Bob pats him on the back. "That's it! What a guy!" And he pats him again.
Ralph looks a little bent, but he is glowing. As a matter of fact, we all are.
After that it was easy. Relatively easy. It wasn't too long be- fore the process was written clearly on the board:
STEP 1. Identify the system's bottlenecks.
(After all it wasn't too difficult to identify the oven and
the NCX10 as the bottlenecks of the plant.) STEP 2. Decide how to exploit the bottlenecks.
(That was fun. Realizing that those machines should not
take a lunch break, etc.) STEP 3. Subordinate everything else to the above decision.
(Making sure that everything marches to the tune of the
constraints. The red and green tags.) STEP 4. Elevate the system's bottlenecks.
(Bringing back the old Zmegma, switching back to old,
less "effective" routings...) STEP 5. If, in a previous step, a bottleneck has been broken go
back to step 1.
I look at the board. It's so simple. Plain common sense. I'm wondering, and not for the first time, how come we didn't see it before, when Stacey speaks up.
"Bob is right, we certainly followed this process, and we cy- cled through it more than once-even the nature of the bottle- necks we had to deal with changed."
"What do you mean by the 'nature of the bottlenecks?' " I ask.
"I mean a major change," she says. "You know, something serious like the bottleneck changing from being a machine to being something totally different, like insufficient market de- mand. Each time that we've gone through this five-step cycle the nature of the bottleneck has changed. First the bottlenecks were the oven and the NCX10, then it was the material release system -remember the last time when Jonah was here?-then it was the market, and I'm afraid that very soon it'll be back in production."
"You're right," I say. And then, "It's a little odd to call the market or the system of material release a bottleneck. Why don't we change the word, to..."
"Constraint?" Stacey suggests.
We correct it on the board. Then we just sit there admiring our work.
"What am I going to do to continue the momentum?" I ask Julie.
"Never satisfied, huh?" and then she adds passionately,
307
"Alex, why do you drive yourself so hard? Aren't the five steps that you developed enough of an achievement for one day?"
"Of course it's enough. It's more than enough. Finding the process that everybody is looking for, the way to proceed system- atically on the line of on-going improvement, is quite an achieve- ment. But Julie, I'm talking about something else. How can we continue to improve the plant rapidly?"
"What's the problem? It seems that everything is sailing for- ward quite smoothly."
I sigh, "Not exactly, Julie. I can't push aggressively for more orders because we're afraid that any additional sales will create more bottlenecks and throw us back into the nightmare of expe- diting. On the other hand, I can't ask for a major expansion in hiring or machines; the existing bottom line results don't justify it yet."
"My impatient husband," she laughs. "It looks like you sim- ply have to sit tight and wait until the plant generates enough money to justify more investments. In any event darling, very shortly it will be Donovan's headache. It's about time you allowed others to worry."
"Maybe you're right," I say, not totally convinced.
"Something is wrong," Ralph says after we've made our- selves comfortable. "Something is still missing."
"What?" Bob says aggressively, all geared up to protect our new creation.
"If step 3 is right..." Ralph is speaking very slowly, "if we have to subordinate everything to the decision that we made on the constraint, then..."
"Come on Ralph," Bob says. "What's all this 'if we have to subordinate'? Is there any doubt that we must subordinate the non-constraints to the constraints? What are the schedules that you generate on your computers if not the act of subordinating everything to our decision about the bottlenecks' work?"
"I don't doubt that," Ralph says apologetically. "But when the nature of the constraint has changed, one would expect to see a major change in the way we operate all non-constraints."
"That makes sense," Stacey says encouragingly. "So what is bothering you?"
"I don't recall that we did such changes."
"He's right," Bob says in a low voice. "I don't recall it ei- ther."
"We didn't," I confirm after a while.
"Maybe we should have?" Bob says in a thoughtful voice.
"Let's examine it," I say. And then, "When was the first time the constraint changed?"
"It happened when some green-tag parts started arriving at assembly too late," Stacey says without hesitation. "Remember our fear that new bottlenecks were popping up?"
"Yes," I say. "And then Jonah came and showed us it wasn't new bottlenecks, but that the constraint had shifted to being the way we released work to the plant."
"I still remember the shock," Bob comments, "of restricting the release of material, even though the people had practically nothing else to work on."
"And our fear that 'efficiencies' would drop," Lou com- ments. "In retrospect, I'm amazed that we had the courage to do it."
309