Question
The current edition of Custom Woodworking Business magazine has a case study of a company called JK Concepts. They discuss workflow and work station logic. JK Concepts has apparently moved away from a bench based assembly strategy. It used to be that all the different operations would occur at the same location. In this respect, the workbench became the equivalent of an all purpose scanner-printer-copy machine, capable of doing anything (eventually), but nothing stellar.
Creating individual work stations for the various processes provides a certain clarity. Imagine the optimum bench for installing drawer slides and how different it would be from the optimum bench for assembling the cabinet. The dedicated drawer slide station might have just the accouterments germane to drawer slide installation, whereas the bench for screwing the box parts together might include a nail gun and confirmat screws.
The typical strategy is to launch the cabinet, then gather the tools together as needed for the various processes. The tools need to be collected and put away when done. There is usually some searching involved because the tools are not always where they are supposed to be and there is usually some unnecessary competition because the tools don't necessarily get put away when they should.
The biggest problem with the bench based strategy is the opportunity for randomness. When everything is democratic, it is hard to sustain an optimum set of sequences or to even understand status of completion.
The work station based model seems to fix a lot of these problems. The tools you need are only where they are needed. The completion status of any particular cabinet can be gauged just by location in the bench cycle. This approach probably also helps to sustain the use of best practice methods. It's harder to stray off the reservation when your facilities prescribe a limited scope of activity.
The best configuration would be one where a bench could accommodate two workers. These same two workers could float with the same cabinet through the various stations on the line. A buddy system, if you will. This approach would at least improve training because the mentor can remain productive and available. These stations could be staffed by one or five as demand fluctuates or resources become available.
This is obviously something that would work best with a defined product line and plenty of real estate. Would it be possible to accomplish some of this within a smaller facility? How would you configure the individual work stations? Could they be developed in a way that builds in flexibility? Could our shops be configured like a great big Taylor clamp where the work stations rotate?
Forum Responses
(Business and Management Forum)
From contributor L:
What it appears you are talking about is an assembly line. That can be full blown or bits and parts. We have one area that could best be described as a cell. When there aren't a lot of items to go through it, one man can operate it. When there are more than 50 cases or so, we add people and break it down into smaller operations per man. At that point it might be considered an assembly line. The advantage of an assembly line is both speed and the "push" that it generates. The cell must be configured with all the required tools always there, in use or not. Few people are really good at everything. Specialization has advantages in reduced setup time, training time and responsibility. Shop size and product are going to be major controls.
One of the accepted realities of integrating new people is that the employer will be under water financially until the newbie gets it figured out. The classic work station does not allow for a second participant and, as a consequence, the trainer or mentor usually leaves the station prematurely. If they could stay throughout the whole production cycle, the quality of training would increase and output would not suffer. This would also keep new people from creating their own policy with respect to method or pace.
The second reason for being able to staff with two is for peak demand or otherwise balancing the line. This two person station seems ideal for teaching. This would be the best way for an experienced craftsman to share knowledge (would that they did not perceive this opportunity as decreasing market share). This station would also be ideally set up to drive with just one participant.
I have no space for dedicated workstations, never mind two-man workstations. Thus, I must eliminate the need for much training where I can. I have been delighted with the results of choosing to view the subassemblies that comprise a cabinet as separate independent product lines, and think of the bucket-brigade production runs for each as separate mini-factories. Workers get lined up along sort of a circle of adjustable height benches, with one bench in the center (only 3' or so walkways between), much like the sewing circle someone mentioned the other day, but with a small buffer zone in between workers - these too are fully size flexible, but I prefer room for a max of 3 parts/units. Then each crew member is assigned a small cluster of task elements, according to his ability, so I consider each worker a cell in the line. Often, many cells/task clusters can be arranged to require very little training, and help is usually only a couple arm's lengths away. Furthermore, there seems to be tremendous flexibility in balancing the task groupings, even mid-run, by simply shifting a task element or two to the cell/worker to left or right. Gathering tools is still a hassle, but I plan to soon elevate the tooling-up of the "cells" by means of a series of mini toolboxes, pulled from dedicated drawers. The tools/toolboxes will be sorted and sequenced to suit the number of task clusters, or cells, that I normally like to employ for a given type of run. To switch from a face frames run to a drawers run, for instance, we would simply toss one set of tool boxes in a drawer and pull out another set. The time required should be hardly noticeable.
Also, to help remove excess thinking about the division of tasks into cells, and skill levels, etc., I'm working on a football style playbook that diagrams how to set up each type of run for the number in that day's crew, and cells would be color coded to show where to place workers of various skill levels, as well as outline a series of rules to guide deployment. Sometimes runs need to be divided into a couple of sub-runs if there are not enough skilled guys that day. When a sub-assembly/component or cabinet has been pulled through the line, thus completed, it exits the circle directly into the finishing area or out the door into the enclosed trailer. I really only have space to facilitate one-pc flow, however doing so is what provided the space required for the finishing area!
Please forgive the incorrect terminology and rough methods. I have no formal training in all this so I'm pretty much just making things up as I go, stealing from you guys all the time.
Our thinking is as a small shop (one/two guys), we take a job through all the stations to job completion, or when really busy with larger jobs, we can man each station and run more of a piecework style production. The hard part as a small shop is making them flexible.
Your decisions are also based on observable information. The way you develop sub-assembly is really great. The best way is to think of your company as a series of small factories. In the morning you might be a door company. After lunch you might be a drawer company. With limited resources (space and people) you will always get the best staffing in a bucket brigade format.
Having a strategy for how you run the play will depend on how many people you can put into the department. Sometimes we build doors with one guy. Sometimes we use three. This depends on how much information we have available and how much skill is on deck.
With this bucket brigade you will soon learn how to staff. Subassembly will be dominated and your next problem will become how and where to store the work in process. My first instincts would be to see what you can do vertically. You can't push the walls out, but you can harvest some space above some of the machines.
Final assembly is where you will become the most random. Here it is useful to think of your company as not one that builds kitchens or not one that builds cabinets. At this phase you are a company that amalgamates parts into parcels that get carried onto a truck.
The very last thing you do takes place on the very last bench. This is the real estate that you want to be the most productive.
Things you can work on that are not real estate driven are information flow. That is the same crisis whether or not you have 1000 feet or 10,000 feet. And remember that this information flow starts at the customer.
A more concrete example might be all the tools needed for cutting an electric outlet box into a cabinet end panel. You need something to locate the center, a pattern to trace, two different drill bits, and a jigsaw. The jigsaw blade needs to be a thin one. It would be great if it was a sharp one and you didn't have to hunt for it each time you needed it. This may seem like a small minded way of approaching your business, but at the end of the day your cabinets are built with a series of small activity steps.
A woodworking company in Idaho that makes (or used to make) hotel furniture did a similar study. Before consolidating work stations it used to take them 4 - 6 weeks to ship a project. After consolidation the same work order took 3 - 4 days. One of the metrics they discussed was travel time. A typical work order now travels 1000 lineal feet whereas it used to travel 1.3 miles. Consolidation reduced travel time to 15% of what it was.
This may seem like picayune cost to pay attention to until you realize the significance on cash flow. If you can build it quicker, you can get paid quicker.
From that moment about 2 1/2 years ago, I started a personal crusade. No more rubbing wood by candlelight; I determined to become a businessman, or get out and sell snow cones or something. I am a long, long way from the mark still. But after stealing from all of you for years and reading over 100 books (mostly used - only about $800 spent!), compiling 2000 pages of hand-written notes, and 20 binders of rough-draft systems and strategies, I feel that at least I have some new ideas to try, to test for results. I have learned that my opinions do not matter - only results. However, by no means have I actually accomplished anything yet.
This may sound a little excessive, but what were my alternatives, really? My hope is that I will come to view this depression, which dropped my sales by 75%, as one of the best things that ever happened to me. Time will tell.
I now believe that powerful opportunity lies dormant within this industry's inherent complexities. In fact, they are a natural barrier to competitors - blue oceans are possible! I am so grateful for all the great ideas shared above.
I'd like to honor all the "givers" on this site by quoting one of you present at this thread today, from a post you made here about 2 years ago: "Making judgments must be eliminated because they require both substantial experience and strong intuition. Rather, design systems requiring only choices."
A kanban inventory of common parts, hardware and sheets, etc., limited to the batch size of a single kitchen, is what I'm really striving to find space for. I suspect this would tremendously smooth production (automatically feed lulls - without conversation!), use up scraps, and serve as additional schedule and financial buffers as well!? Couldn't much of orders prep simply be a quick view of the bins, even before engineering is done? Maybe I can find something less important to put up high, like you said, so I can create kanban inventory storage below.
As far as getting your people to see the light, the other leanish book I have is called "One Small Step Can Change Your Life - the Kaizen Way" by Robert Maurer. It claims we need to feed workers, and ourselves, new methods in small enough doses to bypass subconscious push-back. They/we can change without even knowing it.
This may work to reform their actions, but I don't know about reforming their thinking. In my test runs of bucket brigade, in two instances, I have hired highly skilled, self-employed buddies of mine to join two different production spikes (sort of like mini barn raising events using bucket brigade). In one case, we almost entirely built a full face frame kitchen with four men (including myself). My total cost of labor was about $500.00. This was a feat never before done in my primitive, messy micro-shop. I was shocked when my buddy argued that it made no matter because, after all, "hours is hours - doesn't matter if done in a week or one day." He just participated in something he himself had never done. My hourly income was probably increased four-fold, yet he honestly could see no advantage.
I thought this blindness a fluke until it happened with the other buddy as well. This time we ran a brigade of five workers, not including myself. I mainly served as the lubricator. We actually formed the line on two 8' benches at the CNC shop, and ran 72 boxes straight into the contractor's awaiting trailer (he was installing himself). Done in less than 8 hours, including collating all those parts. I glowed that this test also went extremely well, to me. I utilized two guys with zero training, two skilled, one semi-skilled. My other cabinetmaker buddy that helped knew my cost of labor was about $800.00 and knew that I made almost 5k for the day - a feat still almost unimaginable to me (then rested and reset/prepped, for a week), yet he saw no magic in it, at all. To this day he hasn't even tried bucket brigade or lean methods, even though I try to spoon-feed it to him in love, but he works 7 days a week and is divorcing. I love him, but...go figure.
What first got my attention was a post on WOODWEB. Someone mentioned that as much as 85% of shop motions where non-value adding. Wow. Who would not investigate a claim like that?
I can understand your passion for bucket brigade. Every time we employ it our metrics go through the roof. We are able to plug in completely inexperienced people for certain aspects and they perform as well (actually better sometimes) as people with decades of experience.
You referenced a blue ocean earlier. Blue Ocean Strategies is one of the better books I have read in some time. It's a bit of a slog but very worthwhile once the premise becomes clear. (I wish I would have read the last half first.) Many of the principles in that book apply not only to creating markets but also creating manufacturing paradigms. There really is no reason to benchmark yourself against your competition.
In the end, I am greatly encouraged by those otherwise inclined - that is the opportunity. Can you imagine trying to survive in an environment where certain levels of efficiency became the norm? How many boxes would we have to build to get a paycheck?
This is an approach that could work with any shop, highly custom or factory line. This is how Lean is applicable to any type of shop. No matter what the particulars are for any specific flow there are best ways to apportion the tasks depending on whether or not there are 1, 2 or 3 people available at that moment. Once the line has been balanced for 1, 2 or 3 people there is no reason to change the playbook unless there are attendant changes to technology.
It is not my goal to invest to whip all comers - that would be waste. I only need to be able to tap-out the average competitor in my area to meet my intentions. The potential, though, is almost endless, I think.
But for fun, with a rocket CNC and a good bander, could not the same performance be replicated by simply adding another couple semi-skilled operators to the mix? I don't mean to oversimplify, but that would have cost only an additional $250 or so for the day?
Come to think of it, the overhead, etc. for the shop and equipment and the operators' labor was already factored into the CNC fees I paid. And the overall profit doesn't even really matter because a better shop, such as yours, would probably sell that job for even more.
So, the problem that comes to mind is that the CNC may not be able to optimize the order of cutting to well accommodate a single box flow (my term for this is a "stop-cell"), therefore, the cutting of that job would need to have been done previously, but the day's run could have been picked up at the bander.
Jon Elvrum used to ask the question, what business are you really in? To me the answer was the promotion or PR business. The middle class has no need for this; they require something that looks different than what they have had for too long but doesn't cost too much. The upper class does have a need to promote themselves, retails stores, class A construction, trade shows, banks, restaurants, lawyers offices the same. At any rate, your enthusiasm is a breath of fresh air. I'm sure you will be very successful.
It is a rich offering, so I can see how we could each come away with very different perspectives. If I were to apply the 80/20 rule according to my own impression of what was actionable in the book, I would tender that the idea is to cause the customer to think of my offering as a far, far better value than my competitors', thus making their choice to hire me somewhat of a no-brainer. Beyond a pure pricing play, I feel the authors were showing how these example businesses did it by exploiting multiple variables to create exceptional value. For instance, Circ did it by providing a new, original type of show, while at the same time axing the most expensive aspects of the traditional circus. Moreover, they could provide this new format in many places a regular circus could not go. All these and other advantages added up to create a superior value to the customer, while cutting costs (invented a space for profits) simultaneously, changing the game completely.
I read that book 18 months ago, but I think I got that mostly right. There is not time or space to outline my models, but I found the book to be a very specific guide. I found it fun to employ all the graphical guides and perception maps, etc. I enjoyed cracking into them with the 5 whys exercise they use at Toyota. In fact, I used the 5 why-nots and the 5 hows too. But I did color a lot of it with ideas from elsewhere, so maybe I read too much into it.
The famous TOC maven, Dr. Lisa Lang, spoon feeds the same applications in an extremely expensive (for me) seminar called The Mafia Offer - it is about creating the unrefusable deal. This deal would be one that your primary competitors cannot, or will not, match. She employs lean/TOC and other disciplines in a variety of applications to build in excess capacity, and then give it away as value to the customer in various ways - creating a blue ocean opportunity, although she seems to avoid using the term.
I have worked very hard to develop multiple (theoretical) profit drivers, designed to create synergy. In fact, production efficiency was one of the last things I looked at - working for your money is the hardest way to make it! The use of a lean bucket brigade to price compete in an ugly way is not really my intention, but my math seems to suggest that I could.
There is also a book called "Positioning the Battle for the Mind: A Different Take on Positioning." This is along the lines of 7up positioning themselves against #1 Coca Cola as the "uncola." This automatically positions them and makes them notable against #1. Another one was Avis rental cars: "we're number 2, we try harder."
All of the guys on the forums who are successful have a position in the market and their customers' heads whether they realize it or not. Saying position creates value is tautology.
It just boils down to finding out what is needed and wanted. It sounds like you are already hip to this point. Lateral thinking means that you don't go into this with preconceived notions, fixed ideas, which is deceptively hard, especially the older you get.
