JIT and Y2K?

greenspun.com : LUSENET : TimeBomb 2000 (Y2000) : One Thread

Anyone else notice this?

Seen on Gold Eagle SiteY2K and the Fractional Reserve Economy


-- LM (latemarch@usa.net), April 02, 1999


Thanks, LM, excellent description of the pitfalls of JIT, Y2K or not. Ask yourself what happens if a large sector of the transportation industry goes on strike. . .

-- Old Git (anon@spamproblems.com), April 02, 1999.

What McIntosh says about JIT is kind of correct as far as at goes. Like what is said by a lawyer representing one party in a trial is correct as far as it goes. If you're interested only in collecting anti-JIT propaganda, he's told you all you need to know. Otherwise, a bit more research will pay off.

By removing excess inventory from the entire system, JIT is a technique for speeding up the ol' reflexes throughout the economy. There are many advantages of JIT:

1) As McIntosh says, it reduces costs. Warehouses are expensive, and excess inventory ties up a lot of money.

2) JIT increases production (read:jobs). Demand tends to be fickle. The faster we can respond to changes in demand (not just increases or decreases in demand, but demand for other products), the more efficiently we can meet demand. Just-in-case inventory means we wait for excesses to be sold before replenishing them (perhaps with something else).

3) JIT increases product quality. If a product fails to meet spec or doesn't sell well, we aren't stuck with a whole lot of it. We can implement quick changes (weeks rather than months) to fix problems or meet buying patterns. As companies moved from just-in-case to JIT, they began to use up existing inventory. What they found was that a dismayingly large percentage of that inventory was basically junk - obsolete parts, broken parts, parts that those on the assembly line had been 'tossing back' into the warehouse sometimes over and over for years!

4) JIT permits a greater variety of products and more experimentation. If the cost of trying something new is minimal (no big expensive production runs, no danger of unwanted products gathering dust somewhere), there is much greater incentive to try new and different products. The benefits of hitting a winner are not diminished, while the risk (cost) of a loser is much smaller.

There have been numerous posts on this forum pointing out local retailers who have noticeably increased their supplies of products commonly purchased for y2k preparation -- more shelf space devoted to drinking water, oil and oil lamps, generators, certain foods, etc. Without JIT, this would not have been possible. The producers of these products feel the changes in purchasing patterns within days, and respond accordingly. Again, with just-in-case inventory, the producers wouldn't have noticed for months.

However, the many and varied advantages of JIT cannot be achieved without a high degree of computerization. Serious problems throughout the inventory control, distribution, and ordering systems indeed present more of an immediate hazard with JIT. I think JIT is more robust than McIntosh imples -- that is, the system will fail 'gracefully' rather than instantly. Like a balloon with a slow leak, rather than a sudden burst.

It should also be noted that corporate stockpiling doesn't change the nature of JIT -- these stockpiles represent a change in demand, exactly as though sales were increasing. As stockpiles are consumed later, JIT systems can switch quickly to wherever the demand moves to.

McIntosh is fundamentally incorrect in saying that JIT systems only work well if the demand for a specific product remains relatively constant and predictable. Exactly the opposite is true! The ability to respond quickly and appropriately to changes in demand is the main strength of JIT. If a retailer runs out of something because of sudden demand, JIT allows the system to respond and fill that demand fast.

McIntosh makes this error because he assumes that the warehouses were always full before JIT eliminated them. This is very wrong. When something is backordered today, we wait at most a week or two. McIntosh seems to have forgotton that when backorders happened in the pre-JIT days (and they certainly did), the wait was often many months.

In y2k circumstances, the old just-in-case inventory system had one single advantage: in case of panic buying, the system would require up to week to exhaust supply (if that supply were fully stocked, of course) rather than a few hours. After that, systemic computer failures would allow neither system to replenish any time soon.

Summary: If the means of production and/or distribution break down, an inefficient system is no better than an efficient system. Either is broken. So long as the computers are at least limping, JIT remains superior.

-- Flint (flintc@mindspring.com), April 02, 1999.

er, Flint, I may be wrong on this but I think you've mistated the issue to the extent that if the computers are even limping along JIT is more efficient IF AND ONLY IF ALL OTHER FACTORS (supply, transportation, availability of personnel, etc) REMAIN FUNDAMENTALLY UNCHANGED.

The error in your logic is in assuming all other factors will in fact remain unchanged. Unless all other factors are repaired or replaced immediately upon failure (impossible with degraded JIT) degradation of computer systems, which will degrade JIT response, will also have ripple effects through out all of the other contributing factors.

or did I miss something? Arlin

-- Arlin H. Adams (ahadams@ix.netcom.com), April 02, 1999.


You are essentially correct and raise a good point. JIT and JIC are roughly analogous to the size of the memory cache or disc buffer you have on your computer. A bigger cache just means fewer trips to main memory but increases the probability that lots of stuff in the cash isn't what's really needed. Likewise, if the computer's mother board is faulty or there is a problem with main memory or the disc drive, it won't matter how big or small the buffer is.

-- Nathan (nospam@all.com), April 02, 1999.

When the Soviet Union collapsed at the beginning of this decade, I remember hearing an expert say that one of the principal causes was that their economy had broken down because it had grown too complex to be managed by a centralized planning organization. It had operated OK in the 1930s through 1960s because the Soviets had produced a relatively few items and they were ruthlessly standardized. By the 1980s the Soviets were producing about 10,000,000 individual items. This overwhelmed the planners.

We may be finding ourselves in an analogous situation with JIT and Y2K. We depend on our computers to manage everything today. The invisible hand of the market is still present, but we could not enjoy our wide variety of goods and reasonable prices without JIT and computers. I believe that if the computers fail, we will suffer at least as great a collapse as happened to Russia.

-- Incredulous (ytt000@aol.com), April 02, 1999.

Flint, I did a paper on JIT a few years ago, did research, obviously. I've also posted a couple of things about JIT on this forum, including the thread "Wal-Mart knows what you bought last Christmas." I know that JIT is so effective that Toyota-Japan copied it, for instance--that's high praise. JIT is a superb system.

But I say again--all it takes is a wildcat truckers' strike and within three days (probably much less) supermarket shelves will be cleaned out. Anyone who's seen a supermarket within 12 hours of a hurricane's arrival in the vicinity knows what I mean.

I'm one of those who thinks Y2K is not the only disruption we have to bear in mind; that storing some food and supplies is prudent in any case.

-- Old Git (anon@spamproblems.com), April 02, 1999.

Old Git,

Yes JIT almost cooked our goose.

I run a hospital lab here in the midwest. We rely on JIT to stock our reagants for our machines. Refer' storage is expensive and hazardous to your budget if the refer' fails, thus JIT.

The UPS strike almost did us in. If it wasn't for our priority in line at FEDX we would have been dead.

Point made I hope.

Flint, good post, we tend to forget all the plusses. I however remain skeptical as to how fault tolerant the system is under certain production line kind of industries rather than grocery and discount store examples


-- LM (latemarch@usa.net), April 02, 1999.


I believe I answered your concern at least 3 times in my previous post, and Nathan clarified it a bit as well. JIT needs a little more explanation, I see.

The essence of JIT is *not* the thin pipe. Yes, this was the initial goal, to save on warehousing costs and free up money tied up in inventory. But in practice, the essence of JIT is *short production runs*. Instead of producing a whole lot of identical things all at once, we produce a little bit each of a lot of different things, and do it very often.

As a result, we needed to vastly reduce setup and turnaround times. Production has become far more flexible and adaptable, and JIT caused this. In addition, of course bottlenecks happen pretty often in any system, no matter how computerized. There are still mistakes, accidents, people change their minds. On the manufacturing line, this means that parts often enough aren't *quite* just in time, they're late. Because of our new flexibility, this isn't a problem. We just do short runs of other things while we wait.

This is what I was referring to when I wrote that JIT breaks down gracefully, whereas JIC broke down completely. If computer glitches cause shortages, JIT is better able to adapt to producing what *can* be produced. So JIT really is to our advantage in a degraded production or distribution environment. It is designed to adapt quickly to changing conditions. JIC had much less flexibility, and tended to freeze up in the face of problems.

Yes, a fat pipe will last a bit longer in the case of a complete breakdown. Perhaps you could argue that we could live off what happens to be in warehouses while we fixed on failure. But this seems unlikely. If problems are so severe that production is impossible, it really doesn't matter how big a production run you can't even start. But short of a complete breakdown, JIT is superior.

-- Flint (flintc@mindspring.com), April 02, 1999.

Old Git:

If you've researched this, you know what I'm talking about (I work in a manufacturing environment). A lot depends on the time frame you're talking about.

In good times, little things can cause larger delays in a JIT system, this is true. But not much larger. In bad times, when you're looking at getting the goods in two weeks (as opposed to overnight in good times) rather than two months, JIT starts to look much better. Yes, in a hospital people will die if some supplies are two weeks late. But with JIC, you might get your supplies quickly for a few more weeks, then wait months for the next delivery. Which is worse?

The grocery store example has some problems as well. Highly perishable items have always been JIT - you can't warehouse bread and milk.

-- Flint (flintc@mindspring.com), April 02, 1999.

Flint, I'm curious, where do you live? I ask because I'd like you to give me some insight into how a metropolitan area such as Los Angeles and Southern California in general might fair if there are significant breakdowns. If you aren't familiar with Los Angeles and Southern California then perhaps you can share some insight into how the Hawaiian Islands will do if there are significant breakdowns in shipping, etc?

Mike =============================================================

-- Michael Taylor (mtdesign3@aol.com), April 02, 1999.

Moderation questions? read the FAQ