Lean Manufacturing question

Guys, sorry for another hot topic but...

Can someone please explain to me - in layman's terms - the benefits and the expectations of Lean Manufacturing?
Let me be more specific:

1: What are the expectations regarding delivery times?
2: What are the expectations regarding setup times?
3: Pricing expectations for one or few pieces vs. quantity?
4: Inventory of completed but not yet ordered parts?

These are questions concerning the VENDOR from the OEM-s perspective. In other words if the OEM is strictly adhering to THEIR Lean Manufacturing plan, what can be expected or required from you?

What brought this up now? Have a customer who is a 1st tier supplier to an OEM. He was told by the plant manager, who is responsible for all manufacturing, assembly, inventory and sourcing, that:
1 There can be no inventory of parts or components that are not allocated.
2: Forcasts are for sales and marketing purposes only and are NOT in any way to be used for sourcing or manufacturing.
3: There should be absolutely no cost difference between 1 piece or 100 piece order on the per-unit bases.
4: Setup charges are not allowed on repeat jobs as all the tooling and processes are in existence, therefore there should be 0 time allocated for changeover.

Naturally, these requirements are also effecting my operations as a 2nd tier vendor.

Please don't flame me as I am dead serious about this. I might have even shrugged it off as ludicris(?) if it wasn't for a comment made by a board member earlier this week that sort of falls along the same lines.

Out of context quotes:

""""
Seriously, if you don't know what "lean manufacturing" is or 5s, or if you've never heard of SMED, get some books and start reading.
........

The only thing a machine shop has to sell is machine time. If you spend loads of time floundering around during chage over and prove out and then run really slow cycle times on top of that, you aren't living up to your full billing potential.
........

""""""

Apparently I am missing a real important part of reality, and the technology to allow that reality is completely hidden from me.

As it stands today, I have ionventory of materials, tools to machine them with, programs to run those tools and finished parts in boxes to support the " Wee need 3 extra pieces tomorrow " phonecalls.


So please give me the cliffnotes version of how Lean Manufacturing helps out me, the SUPPLIER? How do I expect the same prices for materials in 1' increments as 12' or 1000 lbs orders? How do I get 5 jobs done in 5 days when the spindle time is 2 days for each? How do I make 1 part for the same unitprice as 100?

Thanks.

Seymour,

Here's a link that might be helpful.

http://www.eng-tips.com/viewthread.cfm?qid=143595

They are implementing it at our plant. Supposedly you make just what the customer orders, and be responsive with quick setup times and Just In Time machining. Works great in theory, but we've dropped from having a 2:1 market share to less than 1:1 with the competition because we now have little inventory and the customers are going somewhere else to get their valves, whereas before we used to build extra inventory in the summer to deal with the winter rush here(oilfield), and could barely keep up in the winter when the rush always came. Now these smart people come up with these figures that it's costing us this much to keep dead inventory per month etc etc.
Funny how people can make all these theories and totally screw up a business. It might work in some industries but it sure doesn't in ours, where the customer can't wait and will go elsewhere.

Dan

You're not missing anything.......it's about as dumb an idea as one could have....


And we wonder why we're getting whipped overseas

Just-In-Time Delivery is for the birds and raises costs. Something has to give, whether it be the end user's costs or the profit margin of the shops. You can't produce parts one at a time as efficiently as you can in numbers. You can't buy material cheaper by the ounce, than you can by the pound.

Some of these college educated idiots that setup these shops like this should have stayed at the dorms baking their brains on the bong, rather than screwing up our world of manufacturing.

Mrainey

Thank you for the link
Apparently I'm not the only one dumbfounded who's thinking that something is wrong with the big picture.

I just had a 9 hour training class on lean manufacturing and here are some of the things I took away from it.

Inventory turns:
Turning your inventory more often means less cash just sitting which is a good thing. Ideally you would only have the material needed on hand at any one time. As a supplier you won't have the perfect scenerio, but if you can turn 5-6 times a year I would think you're off to a good start.

Work flow:
Large batches of parts slow down production in that they take longer to move from one station to the next. Parts in work have costs associated with them based on the amount of labor involved up to the point where they are at. More parts = more cash on the shop floor.

Bottlenecks:
These are from either large batches, or from processes that take longer to complete than others. If the machine can spit out a part every 10 minutes, and processing takes 20 minutes you will have a potential bottleneck. This can point to deficiencies in capacity or any one of a dozen other things.

5S ( aka A place for everything and everything in it's place)
Anyone seen my 3/8 allen wrench? I just set it down 5 minutes ago... I know we've all seen this in action. It's funny how much money that wrench can end up costing the company during its lifetime of use. Shadow boards provide tools when needed and also show which are missing at a glance. Also, most everyone likes to work in a clean environment.

Kitting common jobs:
It's good practice to have all the tools/clamps/bolts/etc. needed for a job in one place instead of always looking for them each time you setup. Time spent walking around is wasted time.

Documentation:
Tribal knowledge is the worst thing a shop can have. If the only person who knows a procedure is sick, on vacation, or even worse moved on you're in for a world of hurt. The more clear the documents the better a new person will do the job. This saves time and money.

Work Cells:
There is no value added in moving a part. You can't sell that time spent, but you will pay for it. Shop organization plays a key part in eliminating obvious waste. The assembly line is a good example.


The overall benefit of lean is for the customer, but when implimented properly will also benefit the supplier. When Boeing went to min/max they sent lean trainers out to get their supply base up to speed as an example. This is some of the basic concepts, and should give you an idea of what the term means and how it can be addressed.

1 There can be no inventory of parts or components that are not allocated.
2: Forcasts are for sales and marketing purposes only and are NOT in any way to be used for sourcing or manufacturing.
3: There should be absolutely no cost difference between 1 piece or 100 piece order on the per-unit bases.
4: Setup charges are not allowed on repeat jobs as all the tooling and processes are in existence, therefore there should be 0 time allocated for changeover.

Click to expand...

Point 1 is good and valid point, parts sitting on the shelf are money not earning interest.
Point 2 . Without forecasting you cannot expect your supply chain to be able to respond to changes in part demand.
Point 3 is also valid
BUT!
Point 4 is where it all starts to fall over

Take one set of parts I make, the customer wants 60 of 1 type and 30 of another per week for the next 5 months
If we used lean manufacturing to make these parts we'd have 4 operations on the turning centres to set up for what would be about a 4 hr run, followed by me setting the big fixture plate up on the VMC in order to have about 2.5 hrs running.
Even though the tooling is sorted and the fixtures/programs are already created, due to the shear time involved:
Turning- Strip current collet + collet holder, replace with large type, install 28 U-drill, O/D rough/finish tools Oil groove tool i/d finish tools, change collet to finish O/d for 2nd op
VMC-Strip off 2 tool-ex vices, replace and clock in big fixture plate,clock in datum hole, set up end mill ,drill,chamf and M8 formtap + the tool proving time for both machines.
You can be looking at 4 hrs setting in order to get 6.5 hrs running, do this every week and thats 4 hrs at $70/hr that your gonna be spending with machines sitting idle not earning you a penny, times that by the 21 weeks left in the contract......

In conclusion, if a customer wants to move to a 'lean' manufacturing method, you must get said customer to give you a base order of how many components per week he expects to be delivered plus give you a signed contract for say 26 weeks (or whatever)

Boris

'lean manufacturing' sounds good in theory but in practise it can cost a heck of a lot more

Calzone

Without addressing the other points, I can assure you that Boeing does have standing PO-s and long-term PO-s definig the total required quantities, scheduled delivery times and the # of units to be delivered at each time.
This flies directly in the face of the typical Lean thinking, as it allows the supplier to decide wether to make the entire qty as a whole or as a batch.
Also, if you look at the link Mrainey has provided, you'll see that large lots DON NOT!!! necessary mean tied-up costs in manufacturing. You , as a corporation MUST look at your costs as a whole and not by selectively tweak the accounting to fit your view.
Large lots seem like a stagnant investment from the viewpoint of immediate cashflow, BUT when you map it over your fiscal year of manufacturing costs, it just might proove to be economically beneficial.
If ordering material as a large lot saved you say 25%, that's cost reduction.
If each setup is 2% of part cost, and let's say you setup only 6 times a year vs. 12, then there is another 10% cost reduction.
If there is no need to reschedule the production line due to unforeseen circumstances that require more parts than what's on inventory, you once again realize cost savings.

Not to say that large lots are always better, but you must do the accounting calculations both ways to find out. And accounting is the name of the game with Lean, as it does not do dick for the end user, nor for the supplier. It only exists for the bottom line.

Boris

Money sitting in inventory does not earn interest.
BUT.
Extra money spent on setup each time the current requirements need to be fulfilled is coming from interest and savings.
Inability to deliver goods on the fly due to lack of inventory is money never earned.

So what's the balance? In strict-Lean terms you don't even calculate the latter two.

What can LEAN do for you? Place your mind in harmony with the universe, and envision an empty wallet. LEAN times are sure to follow

Seems like the more "systems" our industries buy into, the less competitive we become overall as a nation. Consultants can make numbers tell the story the way they want it told, with no regard for the truth. Sadly, the very people most inclined to hire these consultants are the same ones who are too dim to see thru the bull, so they take everything the consultant says as gospel. Anyone ever notice that the training for every one of these magic bullet systems has a great amount of stress placed on the importance of "getting everyone on board"? Simple translation: Don't think logically and use your experience in an attempt to judge whether or not my system is helpful in your particular operation. Just believe and follow blindly because what I'm saying is your ticket to manufacturing nirvana. If the system fails, its because you didn't get on board and implement it properly...... Is it just my warped mind, or do others notice strong similarities to the practices of traveling snake-oil salesmen of yore?

Here's an actual example of a part where I increased the lot quantity recently, and my logic for why it was a good idea to do so. Its a turned part that I sell via mail order for $50/pr. Made the original at the request of a friend who then said he thought there was a market for it, although limited. I didn't figure there was any potential, but thought I'd give it a try since there was no real investment. Turns out I sell about a dozen pair a month, and have for the last couple years.

Previously, I'd make 10 or 12 pair whenever stock was getting low, which usually turned out to be about once a month. The true time spent to setup, run, clean up, etc would run 4 hours on average. $600 worth of parts for 4 hours time and $35 of barstock seemed pretty good to me, so I kept on doing the same monthly routine.

Last run, I had a new 20' stick of CF bar to start sawing, so I decided to saw up the whole thing and run it all into parts. 11 hours later I had 66 pairs of finished parts, or slightly over a 6 month supply. Got bored along the way and made some changes I knew needed to be made but had ignored in the past, so the next run comes up, it should finish in about 8 hrs.

Looking at this from a Lean or JIT perspective, I'm an idiot. I'm carrying $3300 inventory of one part that's going to take till close to the end of this year to sell out. I'm taking up space and spending money in storing the parts and I've lost the future use of some amount of money by tying it up in inventory.

Looking at it from Cliff's redneck logic perspective, and allowing $25/hr for wages and OH on a basic button pusher, I'm carrying $275 labor and $240 material for a total investment of $515 and not $3300. That $515 will turn into $3300 over the next 6 months, for an annualized yield of over 1000% on the original investment. Over the next 6 months I'll also get some value from the 13 hours I won't spend on these parts (6mo x 4hr versus the 11 hours I spent). At a reasonable shop rate, that 13 hours should return another $800, or, put another way, more than pay for all the labor and material in making the 66 pairs of parts, and leave almost $300 additional to pay for the overwhelming warehousing costs of using about 5 sq ft of shelf space to keep the stuff.

Obviously, this isn't any big amount of money even over the course of a year, but its definitely worth doing. OTOH, I can't see for the life of me how it would be anything but a dead loser if I went at it via the one-at-a-time method.

basically lean manufacturing in its simplest form means that you should remove as much waste as possible from the process. Non value added activities eliminated, cut in the most efficient manner cost effective, etc. However, it also means that noone wants to commit to purchase orders and annual volumes so the machine shop can have their material together. Daily deliveries, etc. sound great, but if a part takes 6 weeks to make and they want to get it delivered within 24 hours that means someone has to stand the inventory. Generally Lean is used to push the inventory back onto the next manufacturing tier. Most places I know who do just in time deliveries have local warehousing to the customer. This warehouse has x days of inventory on hand and the customer pulls from there. the machine shop refills the warehouse. The accounting principle is that the customer keeps their money liquid and turns their purchases faster insteading of sinking 1.7 mil into landing gear struts for a 1 month supply all at once.

DUH!

Just give them the one pc price and if they want more - tell tham that you are more than willing to make more for that price!

That's what I have done. What else can you do? Just b/c they are morons doesn't mean that you need to be. Find a way to cash in on their incompetence. But don't tell them!

We have more and more eddycated idiots with ZERO application experience calling the shots.

Also - these companies are NEVER family ran and are corps that are only looking at the short term $ for their stock values. Those of us that build a business to run for the rest of our careers should never play those types of games.

When slow - build stock! Preferably you pick parts that are most likely to be reordered and have a higher labor to material ratio. Choose to leave the more costly material parts for later.

Pay someone to push a broom today so that they can werk O/T tomorrow for your KANBAN? Those outfits are out there.... But for how long 'till they decide that aint werking and they go on to the next "hot ticket" in a few yrs and try to ride that wave?

Think Snow Eh!
Ox

Cliff
You've just ruined it for any good o'l boy looking for employment in the JIT implementation field.

Willie

That's exactly my point. The concept MAY have financial benefits, but only to the OEM and only when properly analyzed (see Cliff's logic ) .
The problem is that the pushing back of inventory and logistical management to the next lower tier is expected to be done at 0 finacial impact. It cannot be done, not without the next lower tier paying the price.
Naturally that cost is often passed back on to the OEM by increased prices/unit.
That however carries a double whammy.
First, since the increased cost is hidden in the part cost, the Lean-implementers can make their numbers work and keep selling the idea to more and more companies.
Second, after a while these increased costs start to look too high, causing the OEM to look for other sources, often sources outside of the country. Not only speaking of US, but the EU as well.

So again, who is benefitting really?
The OEM doesn't save any REAL money, just "paper" money.
The supplier doesn't save any, possibly even loose business.
The end-user doesn't get a better product, nor does he get a lower price.

So what gives?

If you stick a shim under one end of your machine, it will lean manufacture.

My comments to follow are not to support or oppose any of the previous comments, but just to give a limited example in the auto industry. I am also very much a supporter of the "theory of constraints" bottlenecks.

I have a customer that is tier 1 supplier to a Chrysler final assembly line with a major component. This component can have many variables from one vehicle to the next on the assembly line.

The parts must be in sequential build order as shipped to Chrysler. Thus, these must go into the shipping baskets in the required order. The baskets must go into the truck in the correct order.

As an an aside on a product for Corvette we actually built the sequential variations serially on the line. Generally this is not viable because of tooling changes. The typical thruput is one part every 45 seconds.

The assembly plant that feeds Chrysler is about 200 miles away. My customer has there own trucking fleet.

Chrysler may provide the sequential build information as late as 3 days before product is needed.

To the extent that tooling changes are not needed parts may be batched thru in quantities as small as 20, but more typicaly 50 to several hundred.

We do the gaging on these parts for shims. There are differences in the build characteristics when the part number changes. By use of a thumbwheel we can change our setup in about 1 to 2 seconds. There may be as many as a hundred different part setups stored in the gage. Some of these may require tooling changes.

This plant only does assembly and therefore there are other plants that do machining and before those casting or other processes. There are also a multitude of outside part suppliers.

There is no great amount of inventory storage space in the plant. This plant may run product 12 to 24 hours ahead of its actual use at Chrysler. So you have trucks bringing components in at one end of the plant and assembled product being trucked out the other end. Maybe there are 5000 to 10,000 units stacked up at the end of the assemblly lines for sequencing. This plant does not just supply one final assembly plant and/or line.

You can see this is very much a JIT process. But still there are various buffers of different sizes that have much longer than a day or so of inventory. Shims are one of those big inventory items that some sizes may sit for a long time.

One of the problems of this JIT operation is that castings travel rapidly from the foundry, to machining, to assembly. Thus, they do not get the normal ageing stress relief that existed prior to JIT.

A down side is that if you shut down Chrysler for an hour you have about a $300,000 penalty.

Overall the JIT seems to work well. Keep in mind that within an auto plant the vehicles flowing down a line may be different from one to another. For example you may have a red Navigator with air suspension one engine and transmission followed by an Expedition with spring suspension, and a different engine and transmission. Obviously the trim and other accessories will be different. But I believe the windshields are the same.

.

That IS a good example of a justifiable setup.

And I hope that the stresses and liabilities are compensated for!

Think Snow Eh!
Ox

It's a whole pile of bullsh*t, perfect way to ruin a business as far as I'm concerned. Add some 5S to that and you might as well tell the bank to come and get it.

Parts sitting on a shelf ARE bringing in interest. How so? because when you order your parts in 3,6 or 12 months, the price of material will have gone up again, along with everything else involved in making it. I saw this happen constantly. Here's an example. We need 20 pieces of 13" OD stainless and 5 piece of 10" OD stainless. 13" costs $500ea, 10" costs $400ea. There's 5 extra 13" pieces. 2 weeks later order gets changed for the 10" material, now we need 10. Ok lets get another 5, wow it now costs $600ea because its only qty of 5. That's ok we'll turn down the 13" pieces and we'll save $100 ea, yay we saved $500, we're freaking geniuses(insert pats on the back). 2 weeks later, oh we need 5 pieces of 13", how much? $800 each. Do the math, and add machining time. They could NOT get it thru their heads. Most frustrating job I ever HAD.

When you don't inventory anything. You know what happens? somebody screws up and doesn't order the 1 piece you need and the whole job sits for a month or 2 past due date. That was a constant thing. Or the wrong thing gets ordered, or something was wrong on the drawing, then it has to be done again because we had to have that part on that date, and having it 3 weeks earlier was not acceptable.

I'm so glad I got out of that... Arg.

Common sense has gone right out the door. Now we use "theories" and "hypotheses ".

Hey Seymoure,

You asked for a cliff note version of how it could help you as a supplier. I tryed to show you examples of ways to lean your process from what I have learned. I never said that I totally endorse the whole concept either. I was just trying to help out, and now I wonder if it was worth the time I spent typing it. A Kaizan event may be in order.

edit: And just to be clear, we weren't asked to embrace Lean, it was shoved down our throats. I've learned far too many Japanese words in the last couple of years as a result.

If this ends up as a double post, I'm sorry about my thick fingers.
Anyway...


Calzone, I DO appreciate your comments and thank you for the attempt. I did not in any way ignore what you've said, I just do not get this whole thing.
How is it possible that otherwise successful and savvy businesspeople can't see through the inherent BS embedded in the process as a whole.
Fisrt off, it is ALWAYS done as a go-for-broke process with the here is what you do and like it attitude. Do not question it, as it won't make sense to you anyway, just do it.

I've mentioned the large lot question already.
Inventory issues are ditto.

Workcells. Well, interestingly workcells only make sense in large lot production environments. Anything small would have to mean a single ( or limited ) line of products where the workflow is identical. Example: Saw -> Lathe -> Mill -> Deburr -> assemble -> Package/ship.
If you have a part that does not require the mill operation, well you've just created am unused station smack in the middle of your workcell. Sure, some cases your workcells can be made reconfigurable, but that isn't really feasible in many cases, and prolly very very expensive.

Bottlenecks. Lean thinking may identify bottlenecks, but what will it do as a solution? In a pure manufacturing environment it'll likely result in the purchase of another equipment or a better one. I don't need Lean to tell me that, hands-on involvement in my day to day operation should.

Place for everything is just common sense. For the implementers it means coffemaker in the corner, muffins in the cafeteria, beer is in the executive lounge. This is not a joke. For me, I've learned this a long long time ago. Everything where it belongs and where it makes a good workflow sense, as long as it does not create wasted time going to get it, In that case get a second set, even if it's not used 100% all the time. But that's not very leanish.


Anyway, to put this into the original context, something HAS TO give. My customer was told that price for 1 piece should not be any different than 100 piece.
If there is an immediate need for a single piece, then the time required should not be any different as the sum divided by the # of parts in a lrger order. Translated to English: If 100 parts take 600 minutes to make, then 1 part takes 6 minutes, therefore must be deliverable for the same price in less than 48 hours!!! NO JOKE!!!
They will not release any forecasts, as he needs to make as many and only as many parts as are on the current purchase orders. If there is a sold unit that needs parts, but it's outside of the current mfg schedule, he is not allowed !!! to pre-make parts for them, as they are not yet released for manufacturing. This includes ordering of raw materials.
Ordered but not yet due parts cannot be delivered ahead of schedule, as the system will not allow their inventory. Furthermore, if a PO contains multiple delivery schedules for the same part, he MUST use that as a guide for manufacturing and apply the same "Lean" processes, which means 1 PO, 4 delivery schedule, 4 mfg. batches. Period.

To boot, he will have to demonstrate compliance in order to keep his supplier status.

To me it it doesn't mean a hill of beans just yet, I do whatever I see best, but what will this compliance requirement BS mean to him? What if some other company asks me to do the same later?
What if this becomes a standard course of business?

I hope for the customers sake it is a common, readily available material. Start talking aircraft aluminum alloys in odd designations snd this is a train wreck waiting to happen. I don't know what kind of volumes you are talking here, but there may be a way to work around the red tape.... 4 mfg batches can be run back to back without changeover, merely change the number on the paperwork. do not do a final inspection or something until due date to keep it as " unfinished goods". Bottom line the 2 of you have to do what makes you the most money and deliver as the end user desires.