Streamlining High-Volume Cabinet Production
1. How are you recording employee productivity?
From contributor J:
I would time myself and they would have to be reasonably close to my time per unit.
From contributor T:
I'll skip right to question number 4. The biggest thing we have done to improve productivity is to decrease our batch sizes. Having less work in process at any given time has freed up resources to bear on the cabinets we do have in process. Producing parts before we need them just means we are going to need a place to store them and we could have used that real estate for something that made us money.
Do you ever notice how sometimes at the beginning of the job, you don't quite have enough work for everybody, but at the end of the job, you need to come in on Saturday? This is because you've squandered too many resources (non-productive hours) at the beginning and you've run out of budget (not enough hours left) at the end. You are out of balance.
Some people will tell you that time studies are a waste of effort. These are the same people who advocate for batch processing. Batch processing seems easier administratively because there's less to keep track of on the floor. You just know that Cabinet 6 is cut out because all the cabinets are cut out. To minimize your management strokes, you end up overwhelming your physical plant.
The problem is that while you are busy cutting out cabinet parts, you are not hanging doors. Rather than batch milling on Monday, assembly on Tuesday, door hanging on Wednesday, try getting a cabinet into the door hanging department on Monday afternoon. This approach will help keep your cut-out department from being overwhelmed while your door hanging department is starving for work.
In order to get the doors to show up in the door department when they are needed, you need to know how soon they will be needed. To do this you will need to know your process times.
Before you embark on this, you need to standardize your work methods. It doesn't do any good to record process times if everybody has their own way to do something. You need to figure out which is the best way to do something and get everybody to do it that way. Standardizing production methods will give you consistency and this will give you predictability. Predictability will give you the ability to schedule.
The purpose for your time studies is to:
Figure out what the last activity is. (This is the one that keeps you from putting the box on the truck.) Make that activity schedule when to start the preceding activity. The goal is to optimize your production. You never want to build something before it is needed and you never want to wait for anything. The smaller your batch sizes are, the easier it will be to understand the relationship between when something is needed and how long it will take to be ready. Smaller batch sizes will also help you identify which activities add value and which ones don't.
For example: We used to do panel optimizations and spreadsheets on every job. We had so much work in process we had to have these kinds of management systems just to figure out what had been done and what hadn't. We had intricate labeling systems just to figure out what cabinet a part belonged to. All of these strokes cost money but didn't add any value. Since they didn't add any value, the customer didn't give us any money for them. As a result of lowering our batch size, We haven't made a spreadsheet or optimized a sheet good in a long time. We now spend this time adding value to the product. Our jobs now take us substantially less time from start day to ship day. We don't focus on being efficient, we focus on being balanced.
From contributor K:
I think expecting the guys in the shop to build cabinets as fast as you is unrealistic unless they have as much experience as you do. As business owners, we have gone through the phases of learning how to troubleshoot and figure things out as we go, therefore giving us the ability to build cabinets smoothly. We are also the ones who work hours on the design and know everything about the job before it goes into production. Now that is if they are building from start to finish. If you are a production shop and you have guys set up at little posts doing the same thing day in day out, then they should probably be faster than you at some point because they should find a niche and be used to it, whereas you may have to rethink it after sitting in the office for a month.
From contributor J:
Cut out and have the parts ready for about five of each type of cabinet. Put each group together and time yourself. Have your employees watch you do it. This will establish the time to assemble cabinets. Ask their comments on why you are able to do it so much faster than them.
From contributor D:
Our shop is run on the 32mm system. Our time range for each cabinet is about 20 to 25 minutes from cut out time to finished assembly. We have a system flow and we set up each job to fit the flow pattern of the shop.
1. Cut out
Our trick is that out shop is set up to fit our needs and line rotation for parts to flow to each station and a machine for each station (no back tracking to the same machine). 32mm system is the key and using the line bore to the max for hardware placement.
From contributor T:
Your times look pretty good. What kind of saw do you use for cut out? How much time does the edgebanding part take? How many boxes do you cut out before you start the banding process? You say a box takes 25 minutes from start to end. How long does it take to hang the doors and drawer faces for the same box? If you're willing to share this, I'd like to hear about the time relationships between these processes.
From contributor W:
We would be considered a small to medium size shop here in Canada. We employ 17 production workers producing on average 200 to 250 boxes per week. Our mix is 60% frameless and 40% face frame. We outsource our doors, but complete all finishing in house. We utilize pull-processing techniques, as all work is custom serving about 30 retailers with a 3 week lead time. Our shop is working hard at lean manufacturing. All of our staff is cross-trained to complete at least two other tasks besides their home task. We also schedule staff based on demonstrated capacity. Labour requirements are calculated on previous performance plus a stretch goal. Staff schedules are posted 3 weeks in advance by department. Staffing hours are adjusted to meet production demands. Our shop is divided into three departments, each led by a production coordinator. Production coordinators move recourses within their department to meet production goals. We are currently performing at .425 boxes per labour hour. Our goal for 2006 is to average at .48 boxes per labour hour. Each task has been time measured and each month we explore opportunities to improve productivity and quality. As the saying goes, “if you want to improve anything, measure it.”
From contributor T:
You said you implement pull-processing. Your retailers pull product from your shop. Do you use any kind of signal system to alert the preceding work station that they should start the next item? If the production coordinator was not available, how would they receive this signal? Do you have any kind of physical manifestation that an activity has been performed, i.e. cross it off a list, remove a bean from jar, post it to a database, etc.? How does one department learn where another department is in the cycle?
From contributor W:
When I say pull processing, I mean that cabinets are built to order (already sold to customer), not parts built to inventory. I bet you already knew that, but just wanted to check.
As to your question, each operator has a master production schedule produced weekly. The master schedule outlines the general details such as customer, wood type, finish, number of boxes, shipping date and shipping carrier, etc.
The columns on the master schedule also serve a secondary function. The progress of the job is highlighted as it progresses through the department and the plant. Using this method, the first column on the master schedule doubles for job number and for completion at the panel saw. The second column headers are for ship date and progress through our point to point. Difficult to explain, but I hope you’re getting the picture. As work is completed in each task, it is highlighted on the schedule.
Staffs in different departments use the shipping date as a tool to determine their departments’ priority. Example: the assembly department is expecting the parts for assembly on the master schedule's shipping day minus one. If they are not in queue they are after the finish department as to why not. And so on back through the plant. Each department completes a quality audit on its internal supplier including key quality indicators and on time delivery.
The results from these audits are discussed weekly in coordinator's meetings. We put a lot of responsibility on our coordinators. They must be able to both manage resources within their department and work with internal supplies and internal customers. Did this answer your question?
From contributor T:
This is great stuff! What I was asking about the pull system was more about how you kept work balanced between each department. I can see how a company with 17 workers can staff all the stations. With some cross training you should easily be able to allocate labor as the demands within each department differ.
The one man shop is also kind of easy to understand. This guy is going to do everything himself so it doesn't matter so much what rate, or in what order he does things (as long as there is plenty of space in the shop to absorb work in process).
It's the 5 man shop that confuses me. How do you balance production at each department when you don't have enough manpower to staff each work zone? People are going to have to shift from station to station throughout the day. What is the mechanism to keep people from asking, "How long you gonna be on the saw?"?
We tend to rely too much on tacit agreements or ad hoc meetings and as a consequence there's lots of times when nobody is using the saw and lots of times when we need more saws. The challenges don't seem to get fixed by scale of operation. The twenty person shop can get pretty rational about daily strategy, but it also takes more chickens to feed. The one man shop doesn't have so much overhead to support but he better not call in sick. The five person shop costs $1000 a day to run but can't support a midlevel manager who's only job is to hold a clip board.
If you are willing I would like to know more about your transition to Lean. How long ago did you first start to implement Lean Manufacturing systems? Does everybody in your company understand the principles or do they just take lead from the change agents? Can you quantify any of the benefits from pre-lean to now-lean?
From contributor O:
Two men working on the same box, one man gluing and fastening and the other placing and holding will produce far more cabinet boxes than two men working on their own box. When you get in sync with your partner and all the parts are cut and stacked, you can really burn through some boxes.
From contributor M:
Contributor T, I have a 4 man shop - and you're right - we do occasionally bump heads. I've found the answer in outsourcing and job selection. We outsource our cutting and doors and drawer boxes. We then, as a group, assemble, prep, finish and install the job. Then move to the next one.
It is my job to make sure the parts will all arrive at the best possible time, and that I've left enough time to install the product. We can't take on really small jobs because it won't keep us all busy together. We can't take on huge jobs because it ends up killing our cash flow.
We are very good at this: If one person has to wait to use a machine - he will clean, organize, or perform machine maintenance. This helps work flow immensely. We have a really clean and healthy work environment (which looks great to customers). Shop process for the small shop is a perpetual work in progress, but as long as we can identify and correct the things that slow us down, we will keep our heads above water.
From contributor D:
Our study has been on several different size cabinets from cut out to assembly. Our most important tool is the pre-plan of the job, cutlist, material, and hardware on hand and in their rotation layout.
This afternoon we built 14 boxes in 3 1/2 hours. 2 men started at 4:00 and worked until 7:30. This was from cutout to case assembly, including edgebanding, 32mm line bore, dado ends and bottoms for back panels, and box case assembly. No doors or drawer boxes or drawer fronts at this time, just the cabinet boxes.
Our shop layout and machines for each station is our key. My partner and I both have jobs which we do at different stations. This keeps the parts moving down the line without backtracking to the same machine until final box assembly is complete.
We are a small time shop. We bill out under $110,000.00 per year. The shop is a part time job for us. We both work full time jobs during the day. We work from 4:00 to 7:00 + - each afternoon and 8 hours on Saturday. This makes it very important to be organized and have a good shop layout.
The study based on a 24" base cabinet was as follows.
Tools are very basic: 10" cabinet saw, automatic hot air bander with trim station, line bore machine, table saw for dados, nail gun and cordless drills. Not anything very fancy, but it works for us.
To mount out doors takes us 4.5 to 5 min per door. Drill for hinge and mount in door, screw clip in line bore holes. Drawer fronts take 2 min using a jig for placement working off the bottom and 1 side edge of the front to screw off the drawer box to the front. The drawer box takes us about 15 min to cut out and build.
We are small time and have basic machines, but for now it works and we still enjoy the job and make a little money also. Some day we might be able to buy better machines.
From contributor W:
We have been gradually moving towards lean for about the last 4 years. The first step was a new building. The new building made the transition easier in that machine placement was determined to best accommodate workflow. Our line is a “u” shape and conveyers link some tasks together. See diagram below. The numbered circles indicate quality control points.
We use Cut Planner to optimize our sheet stock on a Giben panel saw. This is the only area where jobs are batched. After the saw, parts are separated into individual jobs. By placing components of the job together, we can change priority at any time at any station. Rush jobs are prioritized through the process and recognized by a work order sheet printed on hot red paper rather than regular white paper. These “hot sheets” are recognized as first priority and move ahead of all other work.
Lean manufacturing has allowed us to reduce labour costs, improve productivity, improve quality, reduce raw stock inventory and finished parts inventory. Our staffs are very aware of our efforts. Some are contributors to the process along for the ride. There is a recognized cultural change in our shop. Reward and recognition have played a big part in moving forward. Some staff resisted change and have left our employment. Communication is still the biggest challenge. We are still improving and learning as a team, but our customers have noticed our efforts and that’s what it’s all about.
Click here for full size image
From contributor T:
I like the flow diagram. It makes it so much easier to understand. What is the INDEX station? You said that you batch at the beam saw then componentize the parts so that you can prioritize a cabinet as needed. Do you not batch at finish, or do you have some way of producing repeatable finish from small batch to small batch? I would think it would be simple to have the other processes catch up with finish because you can add manpower as needed. Finish has to cure and you can't do anything about that.
We're working on systems for communication at our shop, as well. The technology part is easy to get. That just takes a phone call. Developing a pull system of task scheduling with variable demand for a wide range of products is the real challenge.
Some of you guys might not know the difference between a push and a pull system. A push system is where some guy starts the job at the saw and then pushes parts onto the floor. The production schedule is set by the sawyer. A pull system is one that is initiated by the customer who pulls a kitchen from your company. This kitchen in turn pulls cabinets onto the truck. The cabinet pulls doors from wherever you get your doors from and boxes from the box department. The box department pulls hardware from the cart which pulls hardware from the distributor. The point is that the last process dictates the previous process... all the way to the beginning of all of the processes. This is a management system that reduces cost by ensuring that things show up when they are needed, but only in the quantity they are needed.
Try this exercise. Take a 12 inch piece of string and try to push it across your desk. You will see that it is constantly puddling up and you are constantly straightening it out. Now try pulling 10 pieces of string. You will see that the pull system is very succinct and the string goes where you want it to. These physics lessons are very applicable in our shops as well. Just like water follows the path of least resistance, your guys will use whatever method is easiest. If you make it easier to succeed than to fail, you will have more successes than failures. If you try to push a set of processes through your shop, you will spend more time straightening things out than you do pushing.
Contributor W, I really like your red work sheet program. No questions asked. Red is number one. Works for everybody.
Would you like to add information to this article?
Interested in writing or submitting an article?
Have a question about this article?
Have you reviewed the related Knowledge Base areas below?