The Goal

I've ever seen.'

'Do you run a balanced plant?' he asks.

'Well, it's as balanced as we can make it. Of course, we've got some machines sitting idle, but generally that's just outdated equipment. As for people, we've trimmed our capacity as much as we can,' I explain. 'But nobody ever runs a perfectly balanced plant.'

'Funny, I don't know of any balanced plants either,' he says. 'Why do you think it is that nobody after all this time and effort has ever succeeded in running a balanced plant?'

'I can give you a lot of reasons. The number one reason is that conditions are always changing on us,' I say.

'No, actually that isn't the number one reason,' he says.

'Sure it is! Look at the things I have to contend with-my vendors, for example. We'll be in the middle of a hot order and discover that the vendor sent us a bad batch of parts. Or look at all the variables in my work force- absenteeism, people who don't care about quality, employee turnover, you name it. And then there's the market itself. The market is always changing. So it's no wonder we get too much capacity in one area and not enough in another.'

'Alex, the real reason you cannot balance your plant is much more basic than all of those factors you mentioned. All of those are relatively minor.'

'Minor?'

'The real reason is that the closer you come to a balanced plant, the closer you are to bankruptcy.'

'Come on!' I say. 'You've got to be kidding me.'

'Look at this obsession with trimming capacity in terms of

92

the goal,' he says. 'When you lay off people, do you increase sales?'

'No, of course not,' I say.

'Do you reduce your inventory?' he asks.

'No, not by cutting people,' I say. 'What we do by laying off workers is cut our expenses.'

'Yes, exactly,' Jonah says. 'You improve only one measure- ment, operational expense.'

'Isn't that enough?'

'Alex, the goal is not to reduce operational expense by itself. The goal is not to improve one measurement in isolation. The goal is to reduce operational expense and reduce inventory while simultaneously increasing throughput,' says Jonah.

'Fine. I agree with that,' I say. 'But if we reduce expenses, and inventory and throughput stay the same, aren't we better off?'

'Yes, if you do not increase inventory and/or reduce throughput,' he says.

'Okay, right. But balancing capacity doesn't affect either one,' I say.

'Oh? It doesn't? How do you know that?'

'We just said-'

'I didn't say anything of the sort. I asked you. And you as- sumed that if you trim capacity to balance with market demand you won't affect throughput or inventory,' he says. 'But, in fact, that assumption-which is practically universal in the western business world-is totally wrong.'

'How do you know it's wrong?'

'For one thing, there is a mathematical proof which could clearly show that when capacity is trimmed exactly to marketing demands, no more and no less, throughput goes down, while inventory goes through the roof,' he says. 'And because inven- tory goes up, the carrying cost of inventory-which is operational expense-goes up. So it's questionable whether you can even ful- fill the intended reduction in your total operational expense, the one measurement you expected to improve.'

'How can that be?'

'Because of the combinations of two phenomena which are found in every plant,' he says. 'One phenomenon is called 'de- pendent events.' Do you know what I mean by that term? I mean that an event, or a series of events, must take place before an-

93

other can begin... the subsequent event depends upon the ones prior to it. You follow?'

'Yeah, sure,' I say. 'But what's the big deal about that?' 'The big deal occurs when dependent events are in combi- nation with another phenomenon called 'statistical fluctuations,'' he says. 'Do you know what those are?'

I shrug. 'Fluctuations in statistics, right?' 'Let me put it this way,' he says. 'You know that some types of information can be determined precisely. For instance, if we need to know the seating capacity in this restaurant, we can de- termine it precisely by counting the number of chairs at each table.'

He points around the room.

'But there are other kinds of information we cannot pre- cisely predict. Like how long it will take the waiter to bring us our check. Or how long it will take the chef to make an omelet. Or how many eggs the kitchen will need today. These types of infor- mation vary from one instance to the next. They are subject to statistical fluctuations.'

'Yeah, but you can generally get an idea of what all those are going to be based on experience,' I say.

'But only within a range. Last time, the waiter brought the check in five minutes and 42 seconds. The time before it only took two minutes. And today? Who knows? Could be three, four hours,' he says, looking around. 'Where the hell is he?'

'Yeah, but if the chef is doing a banquet and he knows how many people are coming and he knows they're all having om- elets, then he knows how many eggs he's going to need,' I say. 'Exactly?' asks Jonah. 'Suppose he drops one on the floor?' 'Okay, so he has a couple extra.'

'Most of the factors critical to running your plant success- fully cannot be determined precisely ahead of time,' he says.

The arm of the waiter comes between us as he puts the to- taled check on the table. I pull it to my side of the table.

'All right, I agree,' I say. 'But in the case of a worker doing the same job day in, day out, those fluctuations average out over a period of time. Frankly, I can't see what either one of those two phenomena have to do with anything.' Jonah stands up, ready to leave. 'I'm not talking about the one or the other alone,' he says,

94

'but about the effect of the two of them together. Which is what I want you to think about, because I have to go.'

'You're leaving?' I ask.

'I have to,' he says.

'Jonah, you can't just run off like this.'