Want a Cheap "Biden '08" Hat?

As Candidates Quit, Trinkets Pile Up (Free CareerJournal.com article)

Here's an amusing article from the WSJ... sort of old news now, but also a story about old inventory, so maybe my delay in posting this is appropriate.

"...small businesses responsible for the paraphernalia are devising creative ways to unload their excess inventory. First come the heavy discounts -- all "Rudy 2008" merchandise is 75% off -- then more-imaginative steps, such as donating the swag to a homeless shelter or promoting a candidate for an as-yet-unplanned race."

Ah, the waste of inventory! I wonder how many "Thompson 2008" buttons and "Edwards '08" shirts were mass produced in huge batches? I also wonder how much of this was produced in Asia, with slow supply chains?

But those same factors made inventory control difficult. With the whirlwind pace, and no clear front-runner in either party until late in the race, the vendors struggled to calculate what they needed and when they needed it. Some products were made to order, but most items were ordered in bulk to cut costs. And while some of what is left could have collector potential, most dropped in value the moment a candidate ended his campaign.

It's a classic inventory planning problem, trying to balance the costs of excess inventory with the costs of stockouts, not being able to make a sale if a candidate suddenly surged in the polls. I wonder if Obama items were hard to find as he rose in the polls?

Ordering in bulk or having long lead times makes this planning more challenging. Longer lead times and bulk orders increase both the risk of stocking out (can't get enough in time) or the risk of excess inventory (oops, wish I hadn't ordered those last 5000 "Rudy '08" bumper stickers).

Mr. Shirey has been in this predicament before. He was the official vendor of Mr. Kucinich's campaign in 2004 and had 25,000 T-shirts left when Mr. Kucinich exited the race. Mr. Shirey tried to sell them to a car wash and investigated the possibility of bleaching the T-shirts to sell them as plain white shirts. In the end, he sold the Kucinich shirts, which cost him $6 each, to a rag company for eight cents apiece.

At least there's some use for them...

The article also describes how a campaign typically farms out their merchandise sales to a small business, a company who (out of loyalty) typically handles only one candidate. The company has an incentive to avoid the risk of excess inventory, but the campaign has the desire to avoid stockouts. So I wonder if the campaign compensates their retailers for any excess? Seems like that would be fair thing to consider a "campaign expense." That's the type of contract I'd ask for if I were the retailer... but then again, I might not get the contract.

One experienced vendor hedged -- by not putting dates on much of the Mitt Romney merchandise:

Some more-seasoned vendors went into the primary season with a better sense of how to handle inventory. Brian Harlin, owner of the GOP Shoppe, supplied the 2000 Republican National Convention as well as President Bush's first inauguration. This cycle, he received the contract to run Republican Mitt Romney's store.

His way to skirt the uncertainty? Many of the items sold on RomneyShop.com didn't have a date attached to them. So instead of designing a T-shirt with "Romney for President 2008," it simply said "Romney for President." With Mr. Romney rumored to be considering a run in 2012 or 2016, the merchandise retains its value. "This all can be used again," Mr. Harlin said.

Four weeks after Mr. Romney withdrew from the presidential race, Mr. Harlin sent out an email blast declaring "Romney Shop Sale Begins Today!" The message read: "This is a great opportunity to stock up for future campaigns."

I wonder if any of the candidate merchandisers tried to "be Lean" -- finding local suppliers (or a company like American Apparel who produces clothing in L.A.) or finding a way to order in small batches?

Subscribe via RSS | Lean Blog Main Page | Podcast | Message Board

"Stock Up?" Really? For a Penny Increase?

Time to stock up on Forever stamps - Yahoo! News

Sigh, first class postage has gone up from 41 cents to 42 cents. That really doesn't impact me, since I mail maybe two or three things a month (thanks to online bill pay -- something that hasn't always served me well).

The postal service did something interesting... they offered a "forever stamp" that can be used at any point in the future. So, paying 41 cents NOW is a hedge against future cost increases.

But does that mean you should have hoarded stamps and bought in bulk? This reminds me of classic non-Lean purchasing department behavior -- we got a great deal, so we bought a ton of them. OK, at least stamps can never go obsolete, so that risk is eliminated here with stamps. And, stamps don't take up much "home warehouse" space in a drawer, so that isn't a huge problem.

Let's assume you mail 20 items per month. That's 240 per year... you'd save a whole $2.40 a year by "stocking up" now (spending $98.40 on stamps). That money, kept in a bank account that earns 3% a year, would earn you $2.95.

What if you bought 30 years worth of stamps? That's a complicated calculation, depending on assumptions of inflation rates, generally and for stamps, over time. This article, from the last increase (from 39 to 41 cents), shows how, historically, the price of stamps has actually GONE DOWN when you consider the value of money over time.

My gut instinct tells me the "stock up" plan probably isn't going to save you much money unless you mail a TON of letters (such as a small business).

Who is benefiting from the "stock up" plan? The United States Postal Service:

The post office sold $267,696,023 in Forever stamps in March, up from
$207,900,132 in February and $115,303,031 in January.

Now, they probably "pulled ahead" sales from future months, but that's a nice cash flow difference for the USPS, getting that money now rather than later.

Don't get me wrong... even if a 30 year buy of stamps doesn't make sense, I'm not encouraging "single piece flow" of stamps. It doesn't make sense, from a total cost standpoint, to drive to the post office (or bank or grocery store) every time you need one stamp. Don't believe someone who says it would be "Lean" to do so, if "Lean" means low inventory. Low inventory of stamps would be traded off by the cost and time involved with driving to the store. Gas... now that's something that's REALLY going up in price more than stamps.

Did anyone stock up on stamps? If so, what was the economic case for doing so? Or did you just buy a pack of 20, as normal?

Subscribe via RSS Lean Blog Main Page Podcast Message Board

Lean Cannot be Measured by Inventory Alone

I know many of you won't have access to this article from the latest Industrial Engineer magazine (from the Institute of Industrial Engineers) but I had to comment on it anyway. The piece, by Richard Shonberger, has the headline "Faltering lean" and has a callout that says "U.S. companies are doing poorly with lean."

At that point, I was drawn in. It's always an interesting topic, the struggles that companies face with lean. There's a whole book on How to Prevent Lean Implementation Failures and I have a separate blog on the topic. Lean gurus and experts are always bemoaning the high failure rates for Lean implementation and surveys are taken and discussed (as we did here). There's no shortage of things to write about and talk about with "Lean failures." The discussion isn't too much different than talking about "Six Sigma failures" or "ERP failures." It often boils down to a lack of leadership and a lack of commitment from the organization, as we've often discussed here.

In his IIE article, Shonberger highlights and focuses exclusively on a single reported financial number, inventory, as his measure of "leanness." I think this is a huge error. I am hesitant to criticize Shonberger, as he deserves much credit for the spread of Lean and Just in Time principles in the U.S. But, this narrow laser focus on inventory numbers does little to help others be successful in their Lean efforts, I believe. It would be like looking at data that shows that teams that win the Super Bowl tend to commit fewer penalties than other teams (I'm making that up, but it could be true) and then assuming that the key to winning the Super Bowl is to avoid committing penalties (and focusing on that almost exclusively as a goal or a metric). A low number of penalties won't necessarily lead to winning (you need to score some points and play some defense, as well).

Lean companies, such as Toyota, Danaher, or others might tend to have to have low inventory, compared to their peers, but low inventory isn't the primary goal of a business. That goal should be long-term profitability. That's how we should gauge the success of a company. Not the short-term profit this quarter, but long-term profitability. Look at Toyota -- the true measure of their success is the sustained profitability that allows them to fund growth and new technologies, creating stability that helps them avoid layoffs and the downward spiral of the layoff cycle.

Shonberger pulled inventory data, what he calls "the measure of merit" (again, I disagree with that assertion) from public companies, something that anyone can do online, and showed that inventory trends are not good.

"Of the 566 U.S. companies tracked, 50 percent show no clear trend in inventory turns and another 15 percent show at least 10 years of worsening turns. That leaves just 35 percent that have maintained a lean trend for at least 10 years or had that trend but faltered in the the most recent 5 to 7 years."

Shonberger continues to make his case by saying:

"Where we see lots of inventory, we conclude, rightly, that the facility is not lean. No other measure is so universal, objective, and available for research."

It might be true that having tons of inventory means you are "not lean" but does having very little inventory on the books prove that you are lean? Remember, Toyota defines TPS/Lean in two parts: eliminating waste (including inventory, as one type of waste) and having respect for people. To me, low inventory, in and of itself, is not enough to prove "leanness."

Shonberger points to Japan and how their inventory numbers have done better recently after 15 years of "malaise" -- and he credits outsourcing. That's a "Lean" approach? It's easy to have huge inventory turns when you don't make anything, if you're some sort of modern virtual manufacturer, the type who only has 10 employees for marketing and design. Is that the path Shonberger wants us going down?

He then points to Toyota and how their inventory turns have fallen from 22.9 (in 1993) to 10.1 (in 2006). But, Toyota is not the same company today as in 1993. It's not an apples-to-apples comparison. Toyota is building and selling more in the U.S. Is Toyota "less Lean" today than in 1993? That seems like a statement that is hard to back up, other than looking at the inventory data. Sure, Toyota has its struggles (defects and recalls), but Lean is only about inventory, right?

So why are companies struggling with Lean? Shonberger points to "weak support in the executive suite," and the temptation to "cherry-pick easy practices" like 5S and kanban instead of focusing on core issues of balancing demand and supply. Is this why Toyota is supposedly struggling with Lean, per the inventory measures? I doubt it.

Shonberger really loses me in his final paragraph when he points to Dell and Wal-Mart as two great Lean examples, he calls them "lean standouts." It's painfully clear we are working off of different definitions of Lean. Dell and Wal-Mart aren't followers of the Toyota model. Dell is just now starting to explore Toyota as a model (as I've complained about before) and Tesco is the clear Lean leader in retailing, not Wal-Mart.

I prefer my definition (shared by many others), which has balanced goals of improving quality, customer satisfaction, employee satisfaction, and company profits. Reducing inventory can contribute to some of those goals, sure. Those are goals that translate across industries. Hospitals aren't trying to get Lean for the sake of cutting inventory levels. That's not a goal that translates and I think it's further proof that Shonberger's definition of Lean is wrong or, at best, outdated.

What do you think?

Subscribe via RSS | Lean Blog Main Page | Podcast | Message Board

Automotive Overproduction Comparison

How GM Handles a Hit: Build Fewer - WSJ.com

Welcome to new readers from the MoneyCentral MSN blog. To explore more about "Lean," visit the main page of my blog or click on "What is Lean?" in the left column.

The WSJ had an article today about how GM is trying to avoid overproduction of hot new vehicles (like the Buick Enclave) so that they don't have to dump inventory to rental fleets or resort to using incentives and discounts to move metal. Both of those practices harm resale value, which is one buying point for many customers.

I read recently how Toyota's goal is to build one car less than customer demand, always keeping that in balance.

So how does Toyota compare to the "Detroit Three" in terms of inventory levels and avoiding overproduction? This graphic from the article tells quite a story.

Toyota has half the inventory of GM, Ford, and Chrysler, not just in total numbers, but in adjusted "inventory per market share point." Toyota carries fewer days of inventory than their competitors, clearly.

Subscribe via RSS | Lean Blog Main Page | Podcast | Message Board

Don't Blame Lean for Part Shortages

Part Shortages Mean Even Longer Delays

First, as a refresher course, here's my take on why Dell is not a Toyota Production System company. The article I linked to above blames "Lean" for part shortages in the laptop industry, this includes Dell.

Some experts are blaming the parts shortages on the unexpected growth of demand for notebook computers around the world. Others are blaming manufacturers such as Dell for keeping only 1-2 weeks of parts inventory on hand ... the "lean" manufacturing policies such as only ordering parts as you need them may indeed have contributed to the delays we're seeing now.

I think the experts who are blaming "Lean" are flat wrong. "Just In Time" inventory is just a part of the Lean methodology. JIT is also *not* built upon the premise of having suppliers half way around the world. With all of their suppliers in Asia, especially compared to ten years ago, Dell can't be expected to get by with such low inventories (and really, this inventory is at their "supplier logistic centers," where inventory is cleverly kept off the books). As supply chains get longer, you're bound to have more variability in your lead times, meaning you have to react by keeping higher "safety stock" levels if you want the same service levels and parts availability. Even though the inventory is off the books, Dell does make certain purchase commitments to their suppliers, meaning that excess safety stock will eventually lead to excess inventory that has to be bled off by offering discounts to customers (hey, we'll give you this 21" monitor instead of the 19" for just a few bucks more).

If this was a true "TPS" system, what would Toyota do if they built PC's in the United States? Would they have done more to keep suppliers nearby in the U.S.? Even going back a decade, we had more parts like motherboards, RAM, and hard drives made here in the country, if I remember right. Would Toyota have done less to push suppliers overseas to cheap labor? Ten years ago, Dell system cases were still produced in Texas, a relatively short supply chain. I believe that's no longer the case. Would Toyota build and ship PC's from Asia to the U.S.? The "final leg" of the supply chain would be longer, but maybe you would have less waste from shorter supply chain legs with the suppliers? Laptops are already built and shipped from Asia (since they're small), but Dell still builds desktop PC's and servers in the U.S. because they're bigger and heavier to ship (and therefore, more costly to ship). Maybe they're suboptimizing that final leg because it's easier to measure the shipping cost to customers than it is to calculate total end-to-end supply chain costs. I hate to think I'm advocating moving PC production overseas, since I helped start up a Dell factory in Austin in 2000...

One other thing that hurts the PC industry site lack of "heijunka" or level loading. To have "just in time" inventory or kanban, every Lean guru will teach you that you first need level production. This helps prevent the "beer game effect" that hurts the suppliers (and makes it harder for them to deliver) given that sales/demand variations are amplified as you go back in the supply chain. Dell traditionally has a huge "hockey stick" effect at the end of each quarter, where sales skyrocket -- not necessarily because of pure customer demand but because sales incentives (for the company, meeting quarter targets, and for individual salespeople) drive demand to be non level. Would Toyota do things to drive sales to be more level so that production could be level and the supply chain wouldn't struggle so much? I'd think so.

So, long story short -- is Lean hurting the PC industry? No. I'd argue its the long supply chains and the lack of level loading that's hurting them. Neither of those things are "Lean" practices at all. To figure out, why do we have part shortages, it requires a few more "whys" than a single why that blames "just in time." Why would you expect "just in time" to work in the PC industry given their other practices?

Final side note, you might remember this post about how Dell had EXCESS inventories reported last year. How is this possible if it's true "just in time?"

Subscribe via RSS | Lean Blog Main Page | Podcast | Message Board

What About The Other Car Makers?

by Dan Markovitz

This blog has commented (here and here) on the WSJ's rush to predict the folly of just-in-time manufacturing in the wake of the earthquake last week that crippled Japanese auto production. ("Blame it on kanban, the just-in-time philosophy of keeping as little inventory on hand as possible.")

What's striking about the WSJ's maligning of just-in-time is that the earthquake forced the temporary shutdown of nearly 70% of Japan's auto production -- not just Toyota's. Honda, Nissan, Mitsubishi, Mazda, Suzuki, and Fuji Heavy Industries (Subaru) also had to stop or slow production last week.

So why attack just-in-time? Why not attack mass production, since it clearly didn't keep the other car companies running? Or the danger of single-source supply for critical parts? Or having factories in Japan, a notoriously earthquake-prone country, for that matter?

Fortunately, Toyota's president, Katsuaki Watanabe, doesn't pay too much attention to the shrill cries of the WSJ. He announced that

the company will examine its risk management and risk control and look for ways to become less dependent on single suppliers. He stressed the auto maker won't change its kanban, or just-in-time, strategy of keeping as little inventory as possible on hand, which reduces warehouse costs and ensures quality.

The solution to these once a decade disasters? Kaizen, of course.

Subscribe via RSS | Lean Blog Main Page |Podcast | Message Board

The Cost of Unsold Uniform Inventory

ESPN - Jersey change too costly for Raiders WR Porter - NFL

Bad news for Oakland Raider Jerry Porter:

The Oakland Raiders receiver said Wednesday he would have to pay $210,000 to switch his uniform from No. 84 to 81 in order to reimburse the team and Reebok for the cost of the unsold jerseys.

I guess with a "leaner" uniform supply chain, they wouldn't have this type of problem? Either way, it's a drag for Porter to get hit with this. You'd think this sort of inventory risk would be part of the cost of doing business for Reebok. What if Porter retired or got hurt, thus reducing uniform sales? Do the Raiders pay a penalty if they had released or traded Porter? I guess you could argue that changing or not changing numbers is truly under Porter's control, but still...

Reducing Waste, Hurting Food Banks

Food Banks Go Hungry - WSJ.com ($$)

This is something I blogged about in August 2005 (with a link then to Joe Ely's blog), now the WSJ had an article yesterday.

Nationwide, food banks -- clearinghouses that distribute food donations to local charitable pantries and emergency shelters -- report receiving fewer donations in the form of imperfectly packaged canned and boxed edibles.

It is the down side of a drive in recent years by manufacturers and retailers for greater supply-chain efficiency. Toward that end, many food manufacturers began producing food in quantities more closely tailored to individual retail customers' needs. That in turn has reduced the amount of food that gets sold to retailers and ultimately returned to the manufacturers.

Instead of relying on excess inventory or misprinted boxes, food banks will have to rely more on direct cash donations so they can buy food. It would be nice if the companies that used to donate, but made improvements, would donate a percentage of that cost savings to the food banks in lieu of the direct donation of defective (but edible) products. Even then, the food banks still might not end up with as much food.

This made me wonder, what came first, the food waste or the food bank?

According to Wikipedia, the first "food bank" was started in Phoenix in 1967. It turns out that the food bank concept was created to handle the waste and excess in the food supply chain (link to St Mary's history):

In his forties, in 1965, van Hengel started volunteering at St. Vincent de Paul, collecting donations for the community dining room. After learning that grocery stores disposed of food that was, either, nearing expiration or had small tears or dents in the packaging, van Hengel began persuading store managers to donate this “edible, but unsalable” food to St. Vincent de Paul. Soon, van Hengel was receiving more food than the one dining room could use.

If the waste had never been there, food banks, in their current form, might not have ever existed.

L.A.M.E.: Thinking Lean is about Low Inventory

IndustryWeek : When Not To Go Lean

Here's another concept of Lean As Misguidedly Executed. I could nitpick the title of the IW piece, from page 51 of the May 2007 issue, for being misleading. For a publication that promotes Lean and always has a few articles about Lean, the headline writer didn't get it, I think.

A quick glance at the headline might make you think, "So Lean Isn't Always a Good Strategy?"

Better headlines for this column, which DOES make very good points, might be:

• "Lean isn't about low inventory"
• "Low inventory might cost you"
• "Losing sales is waste too"
  

The author quotes Professor Marshall Fisher, of Wharton, then adds his comment (in bold):

"If your product lifecycle is short and unpredictable but you have high margins, then overproduction may not necessarily be the most expensive planning mistake you can make." A far bigger mistake, he says, is to lose sales on a full-priced product that turns out to be more popular than you forecast. If you're producing consumer electronics, fashion apparel, books or DVDs, for instance, lean may not be the best way to go.

The author is the one bringing the word "lean" into the discussion, not Dr. Fisher. That is the mistake in the article. Lean is not equal to "zero inventories." Sure, inventory is a form of waste. Having excess inventory is a type of waste. I learned early in my lean career, from a Japanese sensei, that the first goal is to not shut the line down. "Then, low inventory."

Lean is more about total business effectiveness than only one particular metric. If the value of lost sales in a fast-changing or high-fashion industry is higher than the waste from excess and obsolete inventory, then keeping inventory too low might put you out of business.

This is a different dynamic than the auto industry. In 1998, I did a six-month internship at a Kodak division that made semiconductor chips for high-end digital cameras. Before I arrived, someone had the idea that Lean meant getting rid of the buffer inventory between the semiconductor fab and camera assembly. Because the fab (like any fab) had long lead times and highly variable quality yields, they couldn't keep from shutting down camera assembly. The chips were relatively inexpensive and the lack of inventory was keeping Kodak from selling $10,000+ digital cameras to photojournalists. Low inventory was killing them and hurt their business very severely.

My master's thesis (non-printable version linked here or I'll email it to you if you really want to read it) established a framework for determining the right inventory levels to avoid a pendulum swing in the other direction (such as "inventory solves everything.") Not too little inventory, not too much -- the right amount is needed. That's "Lean" to me. In the thesis, I describe how you can't take things like long lead time or variable quality as a given. You have to fix the "root causes" of the need to hold inventory. Inventory itself is a symptom.

Lean isn't just about any one tool (kanban or 5S) or any one goal (low inventory). Lean is a total business system and a management system. It's best to understand the concepts and philosophy rather than just blindly copying the goal of low inventory.

L.A.M.E.: Thinking Lean = "Zero Inventories"

When inventory is not waste - The Manufacturer.com

It's unfortunate that one of the early (1983) popular books about lean (or "just in time," as was the fashion of the day) was titled "Zero Inventories."

Many people who didn't read beyond the title, mistakenly thought lean meant getting rid of your inventory. I've seen this first hand, as my master's internship was spent at a division of a company that thought they had previously "gotten lean" and cut their finished goods inventory to the bone.

Problem was, the production process had hundreds of steps and process yield variation was very unpredictable (semiconductors) ... so they weren't meeting customer demand. It was brutal. They were the bottleneck for a high-tech end product that they couldn't sell enough of at the time. My job was to come up with a methodology for determining what inventory levels SHOULD be, given the lead times and variabilities involved (while using lean to try to drive down the variability, which would allow you to then reduce inventory).

One of my early lean mentors made the point very well. I'm paraphrasing his broken English, but he basically said:

"First flow, then low inventory. Shutting down line is not lean."

That's a lesson that has served me well. Variation shouldn't be a "forever" excuse to hold inventory, but you can just cut inventory without fixing the variation.

I was reminded of all of this reading the linked article, it gives this example:

"... their corporate anorexia may have made them look svelte but not necessarily beautiful in the eyes of customers. Take the example of the Industrial Controls Division of Moog, Inc., which found that, along with its many clear benefits, lean had also produced an embarrassing tendency to miss customer due dates. "

Here's another case where people might say "Lean nearly killed us," but it's more likely another case of "L.A.M.E." (or "Lean As Misguidedly Executed."). How is it "lean" to miss customer due dates? Sheesh.

The article goes on to describe how Moog utilized the services of Professor Mark Spearman, co-author of the wonderful textbook Factory Physics (yes it's worth the price!) and founder of the firm of the same name. Spearman is one of my favorite people in the manufacturing world (he was a professor of mine when I was an Industrial Engineering undergrad at Northwestern) and many of the invaluable lessons of my early manufacturing career are thanks to him.

Spearman's company and software tools allow you to determine "optimal" inventory levels that trade off service levels and inventory costs. I'm sure the tools do the job, but I also assume that Moog, Spearman, and company are not assuming that today's variation HAS to be that way forever. This is the same idea that you can't assume that today's long changeover times ALWAYS have to be that way (thus justifying large batch sizes). Reducing setup time allows you to reduce batch sizes. You don't just arbitrarily cut batch sizes (in many cases) the same way you don't (or shouldn't) arbitrarily cut inventory levels.

On the demand side, working with customers (or salespeople) to help level load sales will reduce demand variation, allowing you to reduce inventory. That's the "heijunka" principle at work.

On the supply side, using TPM tools, for example, will improve your production stability (reducing lead time variation), which will allow you to, again, cut inventory levels.

Just to be clear, I'm not calling Dr. Spearman "lame"!! Companies and managers who make arbitrary decisions without considering the consequences to their customers are L.A.M.E.