Lean/Agile Manufacturing?

Kan Ban, JIT, single piece flow, lean/agile, whatever the favorite flavor of the month is they can work, until they don't. They are production philosophies that some engineers don't have to deal with, depending on the size of your company and how it is structured.

I ran a captive injection mold shop at Motorola when Six Sigma was the latest thing. It's fine most places but when you make 40,000 parts in a single run and one hole is off be .001 inch, what does that mean? Our delivery dates were fluid and we seemed to always be able to recover. But, we worked for the federal government, NASA, NSA, CIA, military where delivery dates were usually not critical.

Your supply chain is your life. If it fails you have problems that can cripple your production. If your product is made in any volume, you could have issues without alternate sources to fall back on.

This is where planning and supplier partnerships come in. This is when you have to pay a premium for performance and delivery. if your supplier/partners aren't up to the task, get rid of them and pay for performance, not promises.

If management is serious and makes a commitment to this, then you are obligated to embrace it and help everyone in the organization to accept and embrace it as their new mantra.

If everyone does not buy in, you will fail.

Good luck.

Agree with lyn. I have to deal with upstream vendors who embrace Kanban and JIT, only it seems many of their upstream components are not supplied IT at all, thus my orders don't make it in time for my customer needs.

In the perfect world where all involved are dedicated to my end result instead of their own, this works very well. However, fuel shortages, tooling issues, labor uprisings, raw material shortages... these can all interrupt the flow of commerce.

2011 Tōhoku earthquake and tsunami is an example of one that ended up hitting some of us in the door industry.

Off by as much as 0.001", eh? I have never figured out why the QC/QA wants a spec like that when the fasteners fitting in the holes do not have that good or better tolerance.

It is really laughable.

I think sometimes, even if everyone "buys in" the entire process is doomed if not controlled and managed very carefully not to be the world's most wasteful time eater.

It's sometimes difficult to educate people about thermal expansion of plastics vs metals.

Specifying an inspection temperature doesn't always sit well with inspectors.

Yes! For plastics, not only temperature, but exposure time and humidity, especially for Nylon and other moisture absorbing materials.

I ran into crap like this once back in my life as a QC/QA guy at a plastic injection molding place here in Lubbock, and they were splitting frog hairs, but fortunately the lab space for that was pretty well environmentally controlled, but they did have issues with variable moisture retention/absorption in nylon parts causing dimensional issues off-spec in some instances. For nylon work, you better have a completely controlled sampling and measurement area, or it is for nothing.

shouldn't take them long to learn...

machining plastics, in a enclosed CNC horizontal lathe, parts can easily shrink 0.030" from when its removed from the warm CNC center and placed on the layout or staging table and cooled.

When implemented based on data AND linked with a continuous improvement methodology such systems are great and lead to ongoing and sustained improvements.

The limiting factor is to recognise where the organisation is now, identify the improvement opportunity, develop a strategy to implement and sustain that improvement and then lock it in before moving on.

The failures usually happen when the improvements are identified and the end condition "expected" without the research and processes put in place to support it.

For instance, if a supplier was able to deliver on 2 days notice once does not mean they can support that under every conceivable situation, it just means you were lucky that time. To then assume 2 day delivery into the future is destined to fail.

However, we were able to get some suppliers down to 2 minute delivery times, by allowing them to maintain inventory in a store on our site. They had responsibility to replenish that store, we had a responsibility to make sure of timely advice whenever we took stock. (Look up "Consignment Stock" as a concept.)

I REALY miss that organisation and the processes that we had developed.

Here at IEEE GlobalSpec we don't manufacture tangible products. We do use Lean principles, at least in my part of the organization. This was in place when I arrived about eight years ago, and for us and for this implementation Lean works really well. Continuous improvement, root cause analysis, eliminating duplicative steps, even viewing the flow of copy from writing through editing and publishing like products moving through a factory -- all of this contributes to efficiency and other benefits.

Introducing Lean, or any other system, to an organization should be handled in pieces, at least in my opinion. Not every system works everywhere.

There is no cookie cutter for this, and if someone thinks they have one, they do not know anything at all about quality systems, or the implementation of them.

Did you tell them that you were already "exceeding their expectations" and the "mission and vision statements" were clearly already "outside the box" and "paradigm shifting?"

You have a lot of info in the answers so far. Most I would tend to agree with. a 'Merry" introduction is likely not the way. If you try to cram something down everyone's throat you are going to get resistance which will likely mean failure.

I too have a small company and "lean" has always been the norm even though no one ever formally introduced it. It developed over the years. I developed procedures and systems as we went, all specifically tailored to my industry and company. What I have developed has kept us in business for a long time in a very volatile and competitive industry.

I think you need to at least give it a chance as you are likely not the only one there that has reservations. It is likely that any "off the shelf" version will not fit your business model so hopefully, senior management is open to some adaptations for you particular business and you can make some suggestions for improvement.

I'm familiar with Lean Manufacturing. The Agile methodology part of it, that I'm reading is that it relates to software development. Lean exposes mistakes so they can be identified and corrected easier, It's also more cost effective not to be carrying inventory into the next year because any inventory carried into the next year is like having to pay rent on that inventory. It's better to just write it off as scrap. Lean and JIT works best if you're located close to your vendors and develop a close relationship with them and make them feel that they are a part of the project. You have a few vendors that can deliver the same materials at your disposal so that you have an better availability of materials and one of them is going to be giving you a better price than the other two because of less overhead costs. If you're a regular buyer then they will project material needs to be able to meet your demands.

Keep in mind the hidden costs like the further away the vendor is from you the more the costs for delivery. You might think you have a good price but you don't see the delivery charge until after the materials are delivered.

It's also important to do constant inventory on raw materials and adjust materials accordingly, especially on high dollar items, because if you're not charging out enough to account for scrap factor, your inventory might reflect having more material than you actually have. If you do inventory once a year for the companies annual financial report, that sudden adjustment takes a lot out of the bottom line and managers yelling "Where's my money?"

Implementation will be your biggest hurdle. The problem is your company culture will need to change.

And for that to happen, everyone (Management), and I mean everyone needs to be on board with it to back up the implementation manager.

And implementing, it has to be structured to your company for the best results, and management has to understand as well as the employees of the company that goals being set.

I wish you luck.

The OP says "merrily implementing". Now exactly what that means is open to interpretation. Any changes should be phased in (and yes structured as you say) giving time for everyone to accept and get used to them. A slam, bam, etc. approach rarely works.

I wasn't kidding when I wished him good luck...

having worked at a company (shipyard) 25 years ago, that was trying to implement a new process such as this... it wasn't accepted through out the company including was senior management who initiated it. And with the poor management we had, when a new management, quality, buzzword process become a fad, they were on it like white on rice... like a cat to a ball of string...

Adopting any current "fad" can be destructive. I am sure that any successful company has systems in place that could be part of any "management" tool but just don't know it by the term in the tool.

Most difficult part for me was the pressure from my customers to adopt all the fads and believe me they try to push them all on you. Fine for the large multinationals who have or hire people to do the paper work, but for very lean (better term might be "well organized and properly staffed") companies the time involved in documenting just to prove you are doing it - similar to ISO standards- is not worth all the time and effort. The actual proof is in the product you ship and the customers acceptance of same.

There are "products" which really do need the documentation and traceability where human life is at risk. Those I am all in favor of, after all, it could be me.

going a little off topic with experience, but still related for implementations,... I agree, like any process, it has a consistent base structure, but you can still make a certain amount of changes to best fit each individual company,... where anyone new to the company, but is familiar with the process can easily adapt.

Definitely, as long it is not "my way or the hiway".

That's when you know you have a poor manager.

Indeed yes.. Had a few of those and took the hiway.

I am a six sigma certified black belt, which includes lean, agile, and several other euphemistic buzz words.

First, if you are seriously going to implement it, it needs to be embraced by senior management, from the top down, and they need to know what they are buying into. That means senior management needs to be trained before you pull the trigger and start implementing. I once refused a job at a company where senior management said they fully supported the effort, but didn't need training because they hired people to do it for them.

Second, plan on 1 to 2 years to PHASE it in. There are a whole bunch of things to put in place and most companies need to really flush out a system as they go. The first main weakness is that companies don't know or document what they are doing NOW and have no yardstick to determine what they want to change. I once worked for a company noted for never being able to do a job the same way once. Seriously. Job one is to document what you do now and most companies don't think baseline data is worth collecting.

Third, adoption of such a system is highly theological. You have to BELIEVE. I also certified as a nuclear quality system auditor and an ISO9000 lead auditor. I found that you can have the best, cleanest, slickest system, but if you don't believe it will work, it will not work. I also saw examples of the most difficult, clunky, arbitrary and unworkable systems that in the hands of those who believed in it, it actually worked. I was surprised to note that about 25 to 30% of the training for six sigma was indoctrination in why you should believe and that you had to believe. I suspect that part of the reason some companies never quite get it is that the professional practitioners of the discipline come across more like Jehova's Witnesses, or,"Hello, my name is Elder Price and I have a book to share with you...."

Personally, I'm Methodist, described as "known to think on occasion", or "a Baptist who can read". I tend to be a skeptic and not a typical rah-rah cheer squad. I appreciate most of the tools that come with the lean/agile tool kit and when properly applied can generate some amazing results. However, you need an open mind and think as you apply the tools, rather than doing the exercises and hoping for a good result. Most of the results of lean/agile initiate at the workfloor/workstation/individual worker level and feed up through the system.

At one point, I put together the regalia for the true six sigma believer. The QPSD hat. The QPSD display shows performance indicators and is located in each work area so that anyone can immediately see how the area is doing. The hat allows anyone to instantly see how you are doing (still crazy). The shadowboard pocket protector. A pen/pencil for everything and each in it's place. You can immediately see if you have a pen/pencil/ruler/caliper/toothpick missing. Finally the place square. This is an 18" x 18" square of clear 5 mil mylar with striping tape around the edges that can be rolled up and carried in a belt holster. In lean, everything has a place and it's marked on the floor with striping tape. The square is your "place". You drop the square and stand on it and you are always in your right place. The piece de resistance was a work instruction for lean pissing and crapping with appropriate floor tape markings, posted laminated work instructions and shadowboard loaded with the proper and necessary tools.

I may not be a believer, but I am an appreciator.

As I understand, Lean Manufacturing insists on minimizing inventory by relying on "Just in Time" deliveries from suppliers. This was a point of contention for the machine shop owner that supplied one company where I worked.

He liked to point out that we didn't hold inventory, he did. Being old-school, he didn't really care. Our demand for parts didn't change significantly month-to-month, so it was easy for him to adjust. He ended up holding two to three months of inventory. Effectively, he said, it just pushes the liability out to the suppliers.

Lean doesn't insist on not holding inventory, it "encourages" holding only what you plan to use in the current "bucket" of production. Ideally the parts magically show up at receiving just as they are drawn for production, referred to as one piece flow.

This really only works where the [parts flowing through are commodity items that can be sourced from any number of suppliers who can be coerced into delivery on a schedule. If the part is custom, then it falls back into the MRP and MRP II type of rules where bills of materials, lead times and minimum or economic order quantities come into play. On these types of items, I have found that JIT usually means just a little late. Usually JIT plans don't include provisions for failure along the supply chain. The best example of this I ever saw was a sawmill in Arkansas that milled orchard grown pine. There was no log deck, there was a line of log trucks going out the front gate and the mill stopped production for lack of logs every time a train came through and the road was blocked at the railroad crossing.

One of the better examples of working towards one piece flow was a situation where an injection molder set up a minifab in the customers receiving area in a cargo container. The minifab received bulk plastic and had dryer, molding machine and trimmer in the container. Finished parts came out the back and went straight into the customers receiving door.

Usually JIT means that inventory is transferred from your warehouse to your supplier's warehouse, particularly if the material moves in batch quantities.

Even in JIT systems, the flow tends to be chunky and divert from one piece flow. The use of Kanban inserts a level of chunkiness because of the restock quantity at the part bin that sits until the bin is empty, the card gets pulled, and the bin gets restocked.

Speaking of the mini-fab, that clearly reduces the load of shipping costs, by merely shipping bulk bins of pellets, rather than finished parts.

I see that as a very clear advantage in unit production cost control.

that's not bad,... holding inventory where you have to pay taxes on it is where the additional costs come in. Hold for that short term is just the costs of doing business

Exactly correct - pushed the liability and the indirect costs back on the supplier. the larger your customer, they more they do that. Insistence on "immediate" delivery, they don't care how, and the they pay you in 90+ days.

Thanks for all of the responses. Much appreciated.

I am still trying to work out the implications of this move to lean/agile, but it appears that I am just going to have to adapt to it. I could move on but the role offers me a tantalising mix of the sort of problems that I like to wrestle with and a nice bunch of people to work with, so better to stay.

I have my criticisms of lean/agile, and I will post on this once I have taken the time to compose it. I intend to point my leader, who is our lean/agile champion, to this thread and so I need to watch my words a bit. He is a nice guy who is trying hard to accommodate this grumpy old engineer.

Later,

Mark aka Paulusgnome

OK, here are a few of my observations. I admit that these might seem a bit superficial, I am trying to avoid writing a novel here.

1. Buzzwords. Sprint, scrum, kan-ban, kaizen … This might seem trivial, but there is a very dishonest psychological trick being used here. You take a word that everyone understands the meaning of, and you give it a twist. A short, highly intense burst of physical effort becomes a normal work spell. A feature of a rugby game becomes a brain-storming exercise. This immediately creates a subconscious confusion around these words in the minds of the participants. Or you use words from another language which also causes confusion in the minds of the participants. These are tactics long favoured by religious proselytisers and others as a means to create believers.

2. Its a religion. You must believe

To me, a key issue is verifiability. If the effectiveness of a particular methodology is not able to be verified and preferably quantified by objective metrics, then requires belief and it becomes the company's religion. As far as I can tell, lean/agile is entirely lacking this kind of verification. I have heard it said 'when I worked at X we implemented lean/agile and got Y% improvement in …' but this kind of anectode falls quite low on the credibility scale for several reasons and only enhances the 'you must believe' feeling.

3.It is teaching us what we already know

Engineers tend to be professional problem solvers who have the mental skills to create good, working tools, methods and processes in response to said problems. This includes the problem domains that lean/agile attempts to claim for itself. Do PEs use scrum/sprint methods? Stripping away the jargon, yes, when the situation suits it. But when the situation demands a fresh approach, its back to basics and dogma goes out the window.

4. While effective in its home problem domains, it must be adapted to achieve effectiveness in others.

Lean/agile was spawned by the IT software industry as a method for managing application software production. It is suited to applications where the product is intangible, the processes are largely limited to human-computer interaction, and the project can be easily subdivided between team members.

When the project includes real, tangible products, or physical processes, or where the project cannot be split into parallel streams, lean/agile methodology steps outside the limits of its origins and has to be adapted to fit.

5. It encourages micro-focus and diverts focus away from critical problems.

Which is exactly what would be wanted in a software coding environment: there are no critical external interrupts and you want the coders to micro-focus on coding. A factory engineer, however, should always be looking for that critical failure BEFORE it becomes a problem and works out a contingency plan. Micro-focussing on that great new product design while the factory operation stutters, ancient items of plant near failure, and spare parts become unobtainable, these may adversely affect operational viability.

6. It makes enterprise-wide critical failure more likely by over-leaning.

Lean preaches the benefits of leaning out the inventory. This is fine when all of your suppliers are lean and reliable, but alas very few actually are. The consequences of stock outages are well understood, but when lean is applied it places some tension between the desire to maintain buffer stock and the desire to minimise inventory. If too much emphasis is placed on lean, outages are more likely to occur.

7. Everything lean/agile? I don't think so. (Thanks Gavin Rossdale)

Lean provides a set of principles and concepts and management methods derived from the Japanese Toyota experience, incorporates a whole pile of Japanese terms, and comes with the 'you must believe' qualifier.

Agile build on lean with methods derived from software application production, add more buzzwords and also comes with the 'you must believe' qualifier.

In reality, both of these systems have something to offer. They are worth studying with a cautious, critical mindset because there are some very effective techniques and ideas embedded in both that could be added to the engineer's project-management toolbox.

But everything lean/agile? Sorry, that would be stretching it a bit too far.

I truly feel sad that you seem so negative about the opportunities that lean can provide.

Yes, there are buzzwords and those "selling their professional skills introduce that side to the implementation, but I've been through a good "lean" experience and the outcome was spectacular.

The foundation principal was to recognise where we were in skills, inventory, supply chain and so on and set that as a foundation to move forward from. Even this first process of "standarise" the work methods created improvement, as it allowed us to select the "best in org" method from those we already had.

As the lean process identified an opportunity for improvement, it was tested, proved and then implemented. For implemented, read standardised across the entire organisation. Once that method became a "way of life" you then move onto another objective.

This doesn't mean that the first improvement in any area is the last. for our manufacturing site one measure was "stockturns" or the number of times the inventory held on site was turned over each year. We thought we were good when we started at around 12 (meaning that on average our raw material inventory averaged around a month.) but when I left we were approaching 200, but we got there by changes each year. Our suppliers learned this new way along with us. Some still chose to run batch production while others moved to smaller lot sizes.

As an engineer in the design group, we would use our inventory system to select from currently available materials rather than introduce another item to inventory. Where possible we would eliminate fasteners and thus reduce assembly time. We would design components to be "self orienting" during assembly.

You talk about it being a "faith", well the shop owners in our town said they could recognise emloyees by how they organised their shopping, or negotiated to purchase a new car. We had benefits outside of work that cannot be measured.

I'm surprised that the first foreign word wasn't muta. The underlying motivation for Lean is the elimination of muta, waste, whether it is time, materials, motions, etc. If you haven't been introduced to Tim Underwood, you haven't been introduced to Lean.

Seriously, though, nothing in Lean or 6 Sigma is new or revolutionary. Everything used has existed decades before being repackaged under newer lables. I'm older than 6 Sigma, and it has been interesting watching its own people change their stories as time goes on.

However, they are all good toolboxes. But you only use the tools you need. You don't pull every wrench and screwdriver out of your toolbox on any given project.

There is one other psychological feature of six sigma/lean. The program is at heart iterative in nature and looks for incremental improvements. Kaizen is the process for making continuous improvements, sometimes referred to in the business as small changes. Kaikaku is another term meaning radical change, or crazy change and is largely ignored by six sigma and lean. Six sigma/lean is typically focused at stable products and processes in a mature commodity industry. The Japanese who adopted and developed these systems are masters at iterative improvement, which is why Japanese commodities so disrupted our purchasing patterns in the '80s. The Japanese as a culture do not do at all well with radical change and this comes through clearly in the methodology. Toolsets in six sigma/lean depend heavily on established process, work centers and equipment, objective evidence, statistical evaluation, consistent normalcy and standard training. One interesting observation in the lean trade is that Toyota did a superior job of adopting lean into their manufacturing, but totally failed to take the process into their marketing and sales.

Agile is another can-o-worms involved in reducing response time to market changes and can take six sigma/lean processes into marketing and management systems. The element of radical market condition change still throws six sigma/lean into a tizzy. (How do you systematically iterate and improve a system that may appear and disappear with the stroke of a regulator's pen or a Twitter campaign?) (Expect the unexpected.)

At this point there are separate six sigma/lean certifications for manufacturing, financial industry and for the health care industry. Mine was in manufacturing.