Getting Started with Lean Manufacturing

Getting the waste out of your workflow is best done in small steps with quick payoffs ó especially in the early stages. February 16, 2012

We would like to consider applying lean principals to our manufacturing process. Where is the best place to start? A book? A course? We are a ten employee shop (three office, three installers, four shop floor) that builds a mix of commercial, residential, and store fixtures. We build frameless cabinets and manufacture with a Holzma panel saw, Weeke ptp and Brandt edgebander. We also employ Cut-rite and Cabinetvision. I would sincerely appreciate any advice and any shared experiences with lean in a shop our size and with a diverse product mix such as ours.

Forum Responses
(Business and Management Forum)
From contributor V:
I am not sure how a small shop like yours with a diverse product mix will achieve substantial results from Lean. I still have traditional thinking. If you take care of quality, price and on-time delivery for each customer, your business should make gradual improvement. Reduction of rework and rejections will naturally improve all three measures of customer satisfaction. This can happen when your people have good skills, common sense and motivation and precisely understand customer requirements and machines are in good shape.

Handling the varying mix of diverse products using limited resources in order to meet stringent due dates must be a challenge to you. Quoting a rational due date and finding a right start time for a new job based on the existing and planned workload and the availability and capacity of resources may also be a problem for your shop. Without a reliable workflow prediction, you might not have good sleep when customers are too strict with your promised delivery times. I do not know how far Lean helps in this regard even if you have no rework and no rejections. Lean experts suggest visual management while I suggest the same along with prediction and what-if analysis of workflow for custom manufacturing.

From contributor C:
We started with the work in progress and the house keeping process. Office, shop floor, assembly, vans, and a truck. Then we jumped to the process of getting a job in and out - from the contact and awarding of the job to the completed product. I would like to say that lean never ends and comes from the top down. We have close to the same set-up, except we have a slider. We have massaged a lot of problems with the above mentioned steps. We are down to very small amount of screws to stock, a couple of guides and things like that. I think the housekeeping thing and tool boards are real time winners in the long run.

From Gene Wengert, forum technical advisor:
Contributor V makes excellent points. I might star his approach from a different angle. Ask yourself what it is that makes your business unique or special to the customer. In fact, your up-to-date business plan should have this info. With this info, you can more easily decide if "lean" or any other change is likely to be appropriate for you. Further, you might remember that someone has to have an inventory, so even if you go lean, that approach can make your supplier increase his costs. Bottom line: Contributor V has provided great insight that must be studied carefully for its applicability.

From contributor C:
Contributor V makes great points and we do have our production times down pretty well. I think there is a happy medium. We had four on the floor, two in the office, and two in the field. We couldn't break even to save our lives. We were lost and an e-mail suggested we start slow and to keep the confusion down. It was a tremendous boost in two weeks. Three years later we are one in the office, two on the floor, and one in the field. We have tripled our output. Our costs are way down, and cash flow is awesome. Just the house keeping and the WIP got us to what Prasad was saying so essential, the basics of capacity and production time prediction.

From contributor Z:
Contributor V - in order for your predictive scheduling system to work you need to have predictable inputs. If you don't know how long something takes to produce you don't know when to start it. Your software does not work without this level of control. Lean gives you the tools to actually get to predictability. Standardization is the driver here. Standardized input produces standardized output. Once Lean is in place you can start to think about your software.

From contributor G:
From talking to others about this, a problem is in getting existing workers to buy into lean. I think whether we realize it or not we have a preconceived notion of the Henry Ford top down mentality which gets in the way of implementing lean. Along this line of thinking I think we have a fixed idea that organizing the shop is a one-time thing. I think we grossly underestimate the time that we should put into organization.

The benchmark on all the organizational philosophies is control. When we get upset or irritated it always comes back to where did we lose control, which is the source of the upset. I like the idea on what to work on in that he says you just work on what bugs you. As control is the goal, what is bugging you is where you lose control.

As prediction and control are the same thing if your scheduling is what is keeping you up at night that is what you should work on. I think this is about the culture in the shop and not a one-time event.

From contributor V:
I have suggested an approach that is based on workflow prediction and what-if analysis functionality. The discussion may not be effective if we bring software issues into it.

You said, "If you don't know how long something takes to produce you don't know when to start it. Your software does not work without this level of control." Yes, you are 100% right in the first statement. Moreover, the start time can impact the lead time in simultaneous production of jobs. It is not easy to know how long it takes to produce something when there are diverse jobs in progress with different due dates, priorities, and process & resource requirements. The production lead times are output not the input for any decent scheduling approach. Many MRP systems failed because they wrongly take job waiting times (between successive work centers) as input and compute meaningless job lead times.

Does Lean really give reliable prediction of workflow in custom manufacturing that involves simultaneous production of several diverse jobs? I am not aware of it for the general case until now. When the nature, quantities and process requirements of jobs significantly vary with jobs and product mix keeps changing without prediction, I do not know how standardization can offer dependable prediction of workflow. I am yet to see a custom manufacturer that has fairly good workflow prediction at any time from Lean methodology. The known variation among the nature, process, and resource requirements of jobs make production management difficult in custom manufacturing. Stringent due dates make the problem more challenging.

From contributor C:
I don't believe the Ultimate Solution is found in Lean or any other method. The solution is found in studying workflow and improving on it, production techniques- dowelled only vs. confirmat/dowel, etc. , information from the office, and material supply/acquisition. Some of the Lean or 5S techniques can be applied, but to implement all at once would create major chaos and interrupt production. The quest is to improve production and get the office to feed it properly and build on that.

Our product offering is anywhere from Corian tops for commercial, prefinished millwork packages, to residential or institutional casework. It can be very chaotic to keep the materials straight.

Housekeeping and controlling work in progress allowed us not only to study our process and times in this diverse mess, but stop tripping on off-fall or incomplete projects hogging precious real estate or very desperately needed cash in the tied up material in many stages of production.

I can definitely say better management of all stages of the business comes from the top down. I don't know about Lean but we have learned to stop paying for things clients aren't going to pay extra for and now we count pennies, dollars and dimes. We can run the shop on the small easy jobs, make payroll and a profit to boot. We never have to wait for that great huge job to bail us out or get us upright again. All employees of a shop truly want to be part of a great team and work in a wicked super shop and lean principles can give that to you, so it helps to implement Lean from your actions and your desk to the rest of the company.

From contributor U:
The way to get started on this journey is to get your head around a few concepts. The first thing to recognize is that lean is an abstract model, a goal that you never really attain. Understanding this will make it easier to reconcile apparent contradictions. Eventually what will happen on this journey is that you will have an epiphany that sorts everything out for you. Kind of like the first time you discover you can ride a bicycle.

For me this epiphany was when I realized Mrs. Smith was only the client but the real customer for the widebelt sander was the door trimming station, the real customer for door trimming was edgesanding and the real customer for edgesanding was door hanging. In short your next process is the one that dictates appropriate batch size.

Another thing to realize is that Lean is not about being efficient. This is probably the biggest stumbling block for most people. It was suggested that one of the hard parts was getting buy-in from the crew. Two expressions you hear all the time in this industry are: The "KISS principle" and "If it ain't broke don't fix it". Another great one is: "How long are you going be on that saw?" These are all related.

"If it ain't broke don't fix it" comes from the CAVE people (citizens against virtually everything). James Womack in his book "Lean Thinking" has a chapter called Concrete Heads. In this chapter he divides your crew up into three groups: The first group are the early adopters. These are the change agents that get the concepts real quickly. The second group is the fence sitters. They could go either way on this but need some leadership. The third group is going to resist you. There are a lot of reasons the third group will not buy in so Mr. Womack simply recommends that you shoot them, particularly if they are in a position of management. These people will do everything they can to convince the fence sitters to not buy in. They use words like "flavor of the month" to describe new lean initiatives. They are quick to point out failings in lean (and there will be train wrecks but you just have to put the train back on the track). In a lean culture of continuous improvement when you are done fixing something you "fix it again".

This is much different than the CAVE people's mantra "if it ain't broke don't fix it". The KISS principle comes from a lack of comprehension of how lean works. It's real easy to sum it up as simply being all about "efficiency" but it's really not that at all. There is nothing more efficient than coping rails for all 30 doors before you move to glue up except that the door hanger can only hang one door at a time and you are starving him of doors to hang because you are busy being "efficient". The reason you have to work on the weekend is because you didn't get all the doors hung because you didn't get started hanging doors until Wednesday. Reducing the batch size will get some doors into the hanging department Monday afternoon.

The customer doesn't care how efficient you are. They don't care whether you put the door on the cart and shove it around the shop 100 times. They don't care if you sprinkle pixie dust on it make it happen. The customer only cares about the shape and color of the piece of wood. The only things you do that add value to this piece of wood are the things that add or subtract material. Running the board over the jointer subtracts material and straightens the shape. Putting the sticks on a cart and shoving the cart to the jointer does not change the shape or color so does not add value. Pushing the cart to the jointer is a necessary part of the process but is still regarded as waste, albeit necessary waste.

Every stroke either adds value or it adds waste. Those are the only two choices. The customer is in charge of determining which steps add value. They will give you extra money to straighten the board. They will pay you the same amount of dollars whether you push the cart ten feet or 100 feet. All cart pushing is waste. Some waste is necessary and some is not. Locate the jointer closer to the saw and you can minimize some waste. In general there are eight kinds of waste. Excess motion is one of the wastes. To this list I would add "excess emotion

From contributor S:
Donít over think your changes and donít waste a lot of time developing master plans. Pick small simple solutions and do it immediately. Instead of spending weeks developing a solution spend a few minutes implementing a solution and then tweak it as time goes by. Do not wait to make the changes. Do them immediately and donít be afraid to admit when they did not work.

Making lean improvements tends to open up many more issues, sort of like trying to align a sliding panel saw or radial arm saw. Fixing one thing means three other things need to be fixed. Do not try to apply lean to broad problems like speeding up installation or streamlining the design to production process; you will only get bogged down, especially when starting with lean. Instead pick simple small actions or events to improve.

In my experience information is one of the biggest issues in cabinet shops. I seldom see small or medium sized shops with good documentation, reports, and paperwork. Do you have an actual "Job Order" report? Does it truly have the right information needed by those who rely on i?. When you send a job to the floor what information do your people need and how are they getting it? Are they asking the boss about basic details like the materials, hardware, toe height, scribe details, schedule, profiles used, reveals/overlay?

I can literally hand my employees the stack of reports in the morning and leave the shop. They are all new to the industry, none more than one year experience. None of them could build a cabinet on their own if all I gave was the elevation and plans. These guys are not cabinet makers. They simply follow the very concise reports that tell them what to do. The do not need to waste time asking questions and looking for me. This is also true for custom work. More than half of our work is custom and a lot of it is not even cabinetry. It is all about getting the right information to the right employee in the right context in the right time and place. We do not have any computerized machines so all cutting and boring is done manually and the reports are what make it happen.