The Goal

"How could I not notice. Okay, Alex, I get your point. There are major problems all over. It seems that in our division there is a whole herd of constraints, not just a few."

"I still claim that there are only few constraints. Our division is too complex to have more than a very few independent chains. Lou, don't you realize that everything we mentioned so far is closely co

"How are we going to do that? How are we going to identify the divisional constraints?"

"I don't know," I say. "But if we succeeded in doing it here, in our plant, it must be possible to do in the division."

He thinks about it for a minute and then says, "I don't think

339

so. Here we were lucky. We were dealing with physical con- straints, with bottlenecks, that's easy. But at the divisional level we'll have to deal with measurements, with policies, with proce- dures. Many of them are cast already into behavioral patterns."

"I don't see the difference," I disagree. "Here we had to deal with all of the above. Come to think about it, even here the con- straints were never the machines. Yes, we called and still call the oven and the NCX10 bottlenecks, but if they were true bottle- necks how come we succeeded to squeeze almost twice as much out of them as before? How come we increased throughput so much without buying more capacity?"

"But we changed almost every aspect of how we operate them, and how we operate everything around them."

"That is exactly my point," I say. "What aspect of operation did we change?" Mimicking his voice I answer, "The measure- ments, the policies, the procedures. Many of them were cast into behavioral patterns. Lou, don't you see? The real constraints, even in our plant, were not the machines, they were the policies."

"Yes, I do see. But still there are differences," he says stub- bornly.

"What differences? Name one."

"Alex, what's the use of pushing me to the corner? Don't you see that there must be major differences? If there weren't, how come we don't even have a clue of what the nature of the divi- sional constraint is?"

That stops me dead.

"Sorry. You're right. You know, Lou, maybe we were lucky here. We had physical constraints that helped us to focus our attention, to zoom in on the real policy constraint. That isn't the case in the division. Over there we have excess capacity going through our ears. We have excess engineering resources that we succeed so brilliantly in wasting. I'm sure that there is no lack of markets. We simply don't know how to put our act together to capitalize on what we have."

Pacified he says, "That brings us to the real question, how does one go about identifying the system's constraint? How can we zoom in on the most devastating erroneous policies. Or, to use your term, how does one go about identifying the core problem, the one that is responsible for the existence of so many undesir- able effects?"

"Yes," I agree, "That's the question, no doubt."

340

Looking at the board I add, "What's written here is still valid. Identifying the system's constraint is the first step. What we now understand is that it also translates into a mandatory de- mand for a technique by which to do it. Lou, that's it. We found it."

The excitement causes me to stand up. "Here it is," I an- nounce, "here is the answer to Jonah's question. I'm going to call him right now. You can imagine my first sentence: Jonah, I want you to teach me how to identify the core problem."

As I turn to leave I hear Lou, "Alex, I think that it might be a little premature."

"Why?" I ask, my hand on the doorknob. "Do you have any doubt that that is what I must learn first?"

"No," he says. "On that I'm quite convinced. I just think that maybe you should ask for more. Knowing the core problem ex- actly might be far from sufficient."

"You are right again," I calm down. "It's just that I was look- ing for the answer for so long."

"I understand, believe me, I understand," he smiles.

"Okay Lou." I sit down. "What else do you think I should ask Jonah to teach me?"

"I don't know," he answers. "But if the five steps are valid, maybe what you should ask for are the techniques required to enable us to carry those steps out. We already found the need for one technique, why don't we continue to examine the other four steps?"

"Good idea," I say enthusiastically. "Let's proceed. The sec- ond step is," I read from the board, "decide how to exploit the system's constraints. That doesn't make any sense to me. What is the point in trying to exploit an erroneous policy?"

"It makes sense only if the constraint is physical, but since we do deal with policy constraints, I guess we'd better move to the next one," Lou agrees with me.

"Subordinate everything else to the above decision," I read. "Same reservation. If the constraint is not physical this step is meaningless. The fourth step is, 'Elevate the system's con- straint^).' Hmm, what are we going to do with this one?"

"What's the problem?" Lou asks. "If we identify an errone- ous policy we should elevate it, we should change the policy."

"How lovely. You make it sound so simple," I say sarcasti-

341

cally. "Change the policy! To what? Is it so simple to find a suit- able replacement? Maybe for you, Lou, not for me."

"For me neither," he grins. "I know that cost accounting is erroneous, but that doesn't mean I've completely figured out what to replace it with. Alex, how does one go about correcting an erroneous measurement or any other policy?"

"First, I think that you need the light-bulb idea, the break- through. The management techniques that Jonah talks about must include the ability to trigger such ideas, otherwise those techniques can't be used by mere mortals. You know, Lou, Julie predicted that as I come to it I'll recognize that we are not dealing just with techniques but actually with thinking processes."

"It started to look like it," Lou agrees. "But triggering break- through ideas by itself is not enough. An even bigger obstacle is to verify that this idea really solves all the resulting bad effects."

"Without creating new ones," I add.

"Is it possible at all?" Lou sounds very skeptical.

"It must be, if we want to plan rather than just react." As I talk I find a much better answer. "Yes, Lou, it must be possible. Look what happened to us with our solution of getting more sales. As a direct result of the French order we threw the plant into a very unpleasant two weeks and we killed or at least delayed a good marketing campaign. If we just thought systematically be- fore we implemented it, rather than after the fact, we could have prevented many problems. Don't tell me that it was impossible. All the facts were known to us, we simply didn't have a thinking process that would force and guide us to examine it early in the game."

"What do we change to?" Lou says.

That throws me off balance. "Pardon me?"

"If the first thinking process should lead us to answer the question 'what to change?' the second thinking process should lead us to answer the question 'what to change to?' I can already see the need for a third thinking process."

"Yes, so can I. 'How to cause the change.' " Pointing to the fifth step I add, "with the amount of inertia that we can expect in the division, the last one is probably the most important."