Home appliances, washers, dryers, refrigerators, stuff like that have been in a bit of a short supply.
Those are not as critical as food, but more so than the latest Xbox.
Surely, a lot of the shortages are because workers have not been at work, or their workplace has been disrupted by Covid.
Balancing supply and demand is not just a problem in manufacturing. Stores have the same problem - few stores here at least have much room for stock that is not on the shelves. We have to do it at home too - Is your garage still full of TP or anti-bac? Even under normal circumstances, I often choose to pay a little more for a smaller quantity because I don’t have room to store the multipacs.
My domestic habits could be described as JIT. I’ve tried the Costco bulk thing but it never made sense for our family. I just don’t have the space to keep bulk purchases and our preferences change so much that we don’t want to eat the same food that often. What we have for dinner is usually purchased from the grocery store on the way home from work.
How would that work anyhow? Most smartphones are recycled or resold even after the original purchaser upgrades, so I think the only ones available in quantity are going to be five, six or more years old. And then how is recycling the chips for reuse supposed to work? You’d have to get enough of the same model phone in large enough quantity and then disassemble them to get the chips. And these things are, I think, mostly glued together so disassembly is not simple. I can’t imagine that it makes any sense economically or practically.
Businesses will be answering these questions for themselves in the coming months and years. Should they change? I think the opinions of people who don’t actually run a business will hold little weight with those who actually do run a business, and in the end, I suspect little will change. Systems with little or no slack certainly do tend to falter/fail under extreme circumstances, but they are most efficient under normal circumstances, “normal” being what prevails most of the time.
FWIW, airlines for many years have tried their damndest to fill their planes very full. Problems arise when weather causes flights to be cancelled, because there aren’t many spare seats available on later flights to rebook the stranded passengers. This can be massively inconvenient around Christmas, and yet the practice persists because customers like cheap airfare. If an airline could be more profitable by flying planes half full and charging higher fares while advertising “guaranteed same-day rebook if your flight is cancelled,” we probably would have seen it by now.
There are indeed some critical strategic resources, for example these two:
Similarly, years ago I toured a coal-fired power plant. At one point during said tour, my attention was directed to a goddam mountain of coal; my tour guide told me that it was a 90-day stockpile to protect against supply disruptions, because prolonged power outages, as the people of Texas recently learned, are really devastating.
I’m not sure the newest version of the iPhone or Corvette (or the parts for same) carry that degree of national strategic importance.
Back in the 80s, I worked in a hospital. The heating came from coal-fired boilers and there was a “mountain” of coal in the yard outside.
This was during a period of upheaval in the mining sector and Mrs Thatcher had decreed that all hospitals had to carry three months worth of coal. Prior to that, coal had been delivered weekly and was stored underground.
During the Trump administration, his energy department suggested or advocated for support for coal and nuclear power plants on the theory that unlike natural gas, oil, wind or solar plants, they could stockpile fuel for a good while.
As an aside, I do not think cars necessarily use the highest end chips. I bought a Lexus in 2009 that still came with a tape player and could not play MP3s, which meant aftermarket upgrades.
It will take time for just in time to be back. It will in industries with low profit margins and tough competition. Perhaps it will make businesses a little nicer to their suppliers. Wouldn’t hold my breath. I agree two years to normalizing demand, maybe longer in the US if they work on infrastructure - rather than fall in the common Canadian politician habit of doing nothing except use the phrase “shovels in the ground” a lot.
I think people mis-understand the overall goal of JIT. It’s not about saving money, at least not directly. It’s about decreasing risk. It doesn’t affect profits much or at all, although it may make the company look more profitable since you’re not holding lots of inventory. And adopting JIT wasn’t done because the accountants loved the idea.
IN accounting, inventories are an asset, but for all the operational functions it creates risk. Companies moved in the direction of JIT (not always in unison or with one clear goal) because keeping all that around was expensive and added problems: it takes space and maintenance and manpower to warehouse it; stuff gets lost, damaged, obsolete, and so on. Plus, many companies in the Days of Yore were holding extensive inventory upstream and down - they had piles of supply and also excess inventory in the sales pipeline waiting to be sold or delivered. But when every firm is doing that, you’re effectively spending a measurable percent of your national GDP on product nobody needs.
There are different strategies for JIT, but it often requires more information sharing and/or close partnership with fewer suppliers. It’s much more than just shrugging and saying, “Guess we won’t stock inventory anymore LOL.”
It is a common problem that interaction on the Internet allows keyboard experts the opportunity to express their assured opinions based upon little more than reading a five paragraph Wired article on a topic. Notwithstanding the input from several people who work in semiconductor manufacturing, real time embedded programming, and automotive, the o.p. is highly self-assured in his essential premise based upon assumptions that have been repeatedly illustrated to be mistaken and ill-informed, and it isn’t as if any information from actually knowledgeable people will persuade him otherwise.
Setting aside the other points that have been made regarding the reliability and suitability of using ARM processors designed for mobile device use and general computing for safety-critical RTES applications, another reason that automotive companies wouldn’t want to build their products (which do indeed use a multitude of separate processing systems) utilizing “recycled smartphone chips” is simple logistics; building a product depended upon a varying supply of recycled components can work if you are building some kind of one-off product like, say, an LS V-8 Tesla, or clothing made from upcycled plastics which are in essentially infinite supply, but if you are building a high volume product that is reliant upon a continuous supply of some particular model of processor extracted from phones that people are throwing away, assuring a continuous supply of that processor is not feasible, especially given that the processors would have to be tested and recertified before use. Nor would be it feasible to build a modern automotive electronics system capable of utilizing different processors depending upon availability; the complexity of adapting software every time a change was made would be prohibitive alone.
“Absolutely bonkers” is actually an understatement for how prohibitively difficult, expensive, and unreliable such a scheme would be. In aerospace, where we commonly use “S-rated” (MIL-STD-883 qualified for the orbital and interplanetary space environment by being hardened against charged particle radiation and with fault-tolerant features) microprocessors, they are actually fifteen or twenty year old technology because of how long it takes to get a new design qualified and the consequences of failure are often catastrophic. There are many smallsat (e.g. CubeSat) applications that do, in fact, use smartphone processors and sometimes literally incorporating a smartphone to use the camera and/or accelerometer as part of its sensor suite, but these are amateur or startup applications where fault-tolerance is not crucial and it is assumed that spacecraft lifespan and criticality is low; as such satellite manufacturers have matured and required greater reliability and lifespan for their business applications, they have migrated toward more expensive processors that if not fully space-rated are at least run through some subset of MIL-STD-883.
Returning to just-in-time (JIT) inventory processes, these aren’t some example of the inherent badness of capitalism or any other political statement; it is a necessary practice in the high volume manufacture of extremely complex devices such as automobiles. It is certainly the case that it is done to reduce cost, as the overhead for maintaining both physical warehouse space and inventory control over enough components to build on the order of a million products of different configurations over a 90 day or 120 day period would be prohibitive, but it would also be problematic for a supplier to produce that supply and then shut down a manufacturing line until the next order comes in. By maintaining continuous production and supply the manufacturer assures that suppliers have a continuity of operations, allowing them to keep staff continually employed, maintain parallel lines so they can be shut down for maintenance and realignment without interfering with supply, and be able to support multiple suppliers for the same product (which automotive companies will generally use to pit against one another to get the lowest prices on many components, although not with microprocessors).
JIT throughout is estimated by using past metrics to predict future production needs, which works fine until there is a unpredicted major disruption such as a pandemic which results in a stoppage of production or shipment. However, even if car companies had, say, 90 days of inventory, there would still be inevitable hiccups in ramping up production just because of the complexity of supply chains. JIT practices may have exacerbated the situation but this is fundamentally a problem with just a really complex supply chain supporting a technological product with hundreds of thousands of individual components, all of which are crucial to building the final product. You can build a canoe out of birch bark and lath, but a modern automobile requires a vast number of complicated components, many of which require sophisticated methods of manufacture and high production volumes to be viable. Just the automotive tire alone is the product of millions of person-hours of development and continuous design improvements even though for most people it is simply a black blob of inflated rubber that holds their car off the road surface.
The issue with not having PPE, respirators, and pharmaceuticals for intubation in response to the SARS-CoV-2 pandemic is not one of JIT inventory but rather a failure to maintain a strategic stockpile of these resources despite many decades of epidemiologists warning policymakers that a global respiratory pandemic was inevitable and statistically overdue. (Most epidemiologists were convinced that it would be an avian or swine influenza, but it isn’t as if a betacoronavirus spillover is any surprise given the 2002-4 SARS-CoV(-1) and the 2012 MERS-CoV outbreaks.) However, politicians ignored this threat just as they continue to ignore even more existential threats such as climate disruption, asteroid strike, the potential for global thermonuclear war, and overuse of non-renewable ‘fossil water’ sources for irrigation and industry along with a multitude of other global health issues and problems.
The United States could have had both a ‘strategic reserve’ of protective and medical equipment as well as manufacturers retained to produce protective equipment on short notice but we chose not to because it was more important to spend the money on multitrillion dollar wars in Afghanistan and Iraq against the largely manufactured threats of “terrorism” and “weapons of mass destruction”, despite the fact that the US is the largest purveyor of both by a wide margin. The issue isn’t capitalism (per se) or the failings of JIT inventory practices, but a fundamental misplacement of priorities on what is really important in terms of safety and security. That a microscopic, not-even-alive pathogen could cause vastly more disruption than nineteen malcontents hijacking planes in a once-in-a-generation event of terrorism clearly illustrates that are priorities need some serious reevaluation and realignment to the reality of the actual ‘threat environment’.
Stranger
The iPhone idea is being discussed in GQ, so I’ll reserve my comments for there.
Spoiler: Not the best idea in the world.
I’ve been worried about the impact of disruption on long supply chains for years. If China blockaded Taiwan, we’d be in deep doo doo, for example. But JIT is not going away any time soon. It is not like this is the first disruption.
In 2001 when the market for servers etc. disappeared, contract manufacturers who bought components and didn’t get paid until they sold the finished boards/systems to the OEM got screwed big time. That was kind of a benefit for companies like Sun and HP who at least didn’t get stuck with the inventory. Imagine how much worse it would have been with a bigger inventory.
And now to a subject close to my um heart. My understanding is that a lot of the TP shortage came from the fact that the manufacturers make two varieties - one on big rolls for commercial use, and one on small rolls for home use. As businesses shut down and people stayed home there was decreased demand on the commercial side and increased demand on the home side, and no matter the disruptions from Covid, it was not easy to switch capacity from one to the other. Yes hoarding was an issue, and panic, but there was a real shortage.
oh nevermind
Also for tax purposes it is an asset. You generally don’t get to write off unused inventory as a cost of doing business.
None of those were problems caused or even exacerbated by JIT inventory. All three were huge demand spikes- in the case of toilet paper, it was that the demand changed format - all the large commercial rolls and stuff weren’t able to be repurposed, and there weren’t enough home-sized rolls to go around. In the case of sanitizer and masks, those were just cases of extraordinary demand. Same with lumber - it’s not really JIT related either. Even had companies/distributors carried say… a couple of months of inventory, we’d still have seen the shortages and price rises, just maybe later, and not quite as dramatic.
Where you’re going to see JIT itself causing issues is when there are relatively short-term transient issues- natural disasters, stuff like ships getting crossways in the Suez Canal, etc… where that issue causes a temporary disruption in the supply chain.
I don’t know much about manufacturing, but common sense (that often over-rated concept) tells me that constantly and efficiently building complex devices requires an in-depth understanding of the entire supply train. Having that understanding is really hard. And requires accurately forecasting demand. For just one example I have read about, several years ago a hurricane hit Puerto Rico and a small chemical manufacturing business was damaged. That disrupted the entire world’s production of automobiles. It turned out that to everyone’s surprise that company was the sole manufacturer of a certain coating that automobile manufacturers specify as environmental protection for their chips. Without that coating, the chips didn’t meet specs. Other solutions existed, but hadn’t been tested in the auto industry. Fairly quickly solved, but the fact that there was only one factory in the entire world that produced that material had been missed by the entire industry.
I’m a big Costco shopper because we have plenty of storage space and Costco is just so convenient. Ironically, I was moving to a more just-in-time personal inventory approach after Christmas 2019 because our storage closet was getting a bit full. Two of the things we intentionally reduced our inventory of were toilet paper and bleach. Just in time for the pandemic.
Canada has a Strategic Maple Syrup Reserve. These reserves are about managing price fluctuations in economically significant commodities more than they are critical to, let’s say, national defense.
Technically, we have one but it was entirely inadequate for its purpose during the pandemic (which I’m sure you realize) but not because we spent the money on wars. Rather, most of its limited funding was used to acquire mediocre vaccines for Anthrax at the behest of a Congress still scared of the 2001 Anthrax attacks against them and under intense lobbying by the vaccine’s maker. Good article behind the paywall of the New York Times.
Was it an SC430? That was famously the last new car in America to offer a tape deck. It’s obsolescence was a talking point even then. Lexus hadn’t spent any real money upgrading that model for years and it probably figured its experienced owners were likely to appreciate it anyways.
No, a much more basic model. By this time, most other marques were offering MP3 and it was surprising to be offered premium speakers with such out of date hardware. The tape was neither an asset nor a disadvantage. Maybe I used it a few times. I presume they have given these things higher priority - easy and cheap to do, certainly something consumers notice.
I suspect they were right in this regard.
Isn’t that the one that Top Gear called the worst car sold?
Okay, first of all I think we should acknowledge that the market system worked amazingly well throughout the pandemic. The flow of goods has continued largely unabated despite lockdowns of workers, travel bans, and other disruptions. No one starved, supply disruptions like toilet paper resolved themselves quickly, etc.
This is in part because of JIT inventory practices. Large firms have HUGE supply chains. The last company I worked with had over 50,000 vendors in their supply chain. These supply chains generally have built-in redundancy (multiple paths to the same intermediate products), but one of the risks are hidden dependencies. For example, you source a key component to five different vendors, thinking that gets you lots of redundancy if a vendor goes bankrupt or cancels a contract with you. But it might turn out that each one of those vendors relies on a common source somewhere, and if that common source fails so do all your ‘redundant’ vendors. So a lot of supply-chain management work is done to suss out and eliminate those kinds of hidden dependencies.
JIT also enables better quality control. If you are testing batches from a vendor, if their quality drops you can either shift to another or straighten them out. If you’ve got 5,000 widgets in your own inventory and someone comes along with a cheaper, better widget, or one that’s more reliable, you either have to eat the inventory of old parts or use them up before switching.
There’s also an aspect of division of labor. JIT allows you to source out lots of stuff that you’d otherwise have to make in-house, and you’re probably not as good at it as the people who do nothing but make those specific parts.
Modern supply-chain management is also responsible for innovation and improvements in efficiency. Yes, this can lead to more profit, but it also means fewer materials needed, lower environmental cost, lower energy usage, etc. These are all good things that you might miss if you think “JIT just means more profit for the rich at the expense of risk for everyone else”.
For example, I worked with a company that had an assembly line that would occasionally produce failed boxes for a product. The engineers looked at the pattern of failure, and discovered that their machines were rejecting paper for the boxes that was in spec, but on the high side (they’d spec the paper as 20lb, plus or minus X%). They originally set these specs because of random sampling from all their suppliers showed that that was the variance in thickness of the paper. But once they realized their specs would too loose for the machinery, they sent out a new order for paper with a much tighter spec. Most of the vendors couldn’t handle it, but one figured out how to make their paper more precisely. In so doing, they also realized that they could make the average thickness of their paper less, while still staying in the lower spec. This actually reduced their costs and the amount of pulp they used, and eventually the other manufacturers of paper figured out why they were losing business to the efficient firm, and changed their own processes to match the tighter spec and lowered their own pulp consumption.
The result is that cardboard boxes for goods are now thinner, and the huge amounts of natural resources and energy that used to go into the manufacture of those boxes was reduced significantly. Other companies needing boxes picked up on the change, and the whole industry got more efficient.
That’s the kind of place where real economic growth comes from: Smart people making tiny improvements on the margin under pressure from customers and the competition. JIT enables this by putting vendors in constant, day-to-day competition. In a world of big inventories, customers would just order one huge supply of paper, store it in a warehouse, and then if their machines weren’t accepting the paper they’d just have to fix the machines to be more tolerant of the slop in paper size.
By the way, that factory was so large that we were told that at any point of the day or night there is at least one train somewhere in the country filled with materials headed for that factory. Imagine how much inventory they would have to store for even 30 days of material. Even their working inventory filled a huge warehouse.
What we will see from SCM in the future is probably a stronger push to make sure there are alternatives for single-source products, re-shoring some critical parts offshored to other countries, and more work to make sure the supply chain is robust and not riddled with hidden dependencies or other weaknesses.
But again, the supply chains held up amazingly well during the pandemic. Governments, on the other hand, were a mixed bag and some utterly failed at the most basic of tasks.