Maximizing Efficiency with CNC Cabinet Production

A CNC-equipped cabinetmaker explains how he squeezes the most productive value out of his equipment. September 8, 2007

I own a Weeke bhp 200 CNC. It is a 5x12 machine that is fully loaded including an off loader. I have been disappointed with my performance and would like to know what others of you are getting for performance with your machines. We are a custom shop that does one-at-a-time kitchens. Average job is 40-50 sheets, 4-5 different materials. We mark the parts by hand and put them on carts while the new sheet is being cut. We run our feed rate at 33m/m and router speed at 21,000 rpm. Let me know how long it takes others of you to run a job similar to this.

Forum Responses
(CNC Forum)
From contributor M:
One man can cut, band, and hardware cabinet parts for 32 sheets of 4 x 8 and 7 sheets of ?? material in a ten hour day. Cycle time per sheet of material is 15 minutes. This cycle time will allow for the assembly of approximately 10 cabinets by the same single operator during same single ten hour day.

Feed speed and spindle speed have very little to do with the overall amount of time it takes to produce parts. It's all about cycle time and how the operator is set up to use his time while the router is doing its thing.

You stated that you marked parts by hand and then put them on a cart. Neither the time spent to put parts on a cart nor to hand label the parts accomplish a task that adds any value to your cabinets, and the labor involved is human labor that is expensive. Also the router has much prettier handwriting than me or any operator I've had working for me. We allow the router to route the part number and edgebanding info on each part. The extra time required to label parts is done by the router while the human is sticking parts in the edgebander.

Putting parts on carts to move them to the next station also requires labor. In our case we push the parts off with the router onto a workstation sitting at the edgebander. Now instead of a human putting parts on a cart to take them for a ride, the operator is picking up a part once to stick it in the edgebander.

At the out feed end of the edgebander there is a very small sorting area capable of holding 10 cabinets worth of parts. After banding a sheet worth of parts, the operator is able to quickly gather up parts for a single cabinet and hardware them. If he wants to begin assembling the parts he just edge banded, then he takes two to three steps to the assembly area and puts a box together.

Most boxes cannot be completed within the 15 minute router cycle time. The number of employees in an operation and the box production required for any single given day will determine if you go ahead and ignore the router sitting idle for an additional few minutes while the operator finishes building the box he is assembling before he loads another sheet.

If you use a single operator to do this entire process you will find it is better to complete one box at a time instead of loading another sheet of material when the router is ready. Loading the router every time it is ready causes the number of completed parts to increase in a way so that you start stacking parts up on carts or in a rack while waiting for their turn to be hardwared and assembled. Waiting in line for anything isn't profitable. There's also no profit earned for or in parts being stacked so they can wait for their turn to be assembled.

You can only get so many parts on one cart or in one rack. The minute you have to push one cart or rack out of the way and get another rack to continue stacking parts, you start spending labor dollars that do not accomplish anything of value.

By extending the router's cycle time by 5 minutes to 20 minutes, you might decrease by 8 the number of sheets being cut in a day. The difference will be how many completed boxes you have sitting available for delivery and installation. The number of finished boxes per day or week will determine how much money you put in the bank every week.

The number of sheets of material you cut in a week determines how much money you will have tied up in raw material. Customers don't like to pay for raw material. They like seeing boxes hung on the wall and base cabinets being templated for countertops.

If you choose to put two people on the process described above, this can also be done. You have to be careful not to fall back into the habit of doing things with these two operators that add no value to the process. This can happen because of a choice of construction method, software limitations as to how you process parts, or workstation layout in your plant.

In the past I also owned a Northwood NW58. The cycle time on this machine was down below 10 minutes on a 5 x 8 sheet of material. The decreased cycle time of this machine with one man trying to do everything did not increase the number of completed boxes in any given time. All the decreased cycle time did was increase the work in progress as long as I tried to feed it a sheet of material every time it was able to cut another piece.

Had my box production required more finished boxes, I could have put a second person in this cell and as long as I balanced each operator's responsibilities, then the decreased router cycle time would have helped to increase the number of finished boxes without increasing dollars invested in work in progress.

If you think your requirements are such that you need two operators between cutting, banding, and assembly, then you also have to look at other processes in your operation other than just box production. If these other processes are not balanced to the two man cycle time on the router, then you will not increase the amount of finished product available for delivery in a day\week.

Jacking up box production without balancing resources in other areas of the operation has the same effect as cutting a bunch of parts so you can stack them on a cart so you can eventually get around to banding them. This unbalanced use of resources could be called constipation for the pocketbook.

From contributor S:
Great points. I don't work in a cabinet shop or pretend to know anything. But I wonder how cabinets that will require parts of differing materials should fit into the flow. I agree that a part that is sitting idle is not efficient. But let's say that for some reason we have a box that has three different types of material going into its construction. I would pretty much call something like this a custom box. Would it be better to have a cart dedicated to each custom box and place the parts for that box on that cart as they are cut, and then proceed to assemble, including further machine steps once all of the parts are cut? Or do you put all cut parts into some sort of indexed bin or storage system and then when it is time to assemble the custom box, pull the parts and assemble?

With the nesting of parts on the raw sheet stock and differing materials per box, how do you keep the flow from sheet to box going with minimal part idleness and low part handling? Just curious. I know that I try to set things up for a steady flow, but what often happens is that constant work fools people into thinking that there is constant production.

From contributor M:
One factor I listed that affects the amount of unproductive time in a process is the construction method. Included in the construction method is the choice of materials used for boxes. We have chosen a construction method that causes the need for two different types of material for any box. We use ?? thick material for box parts and ?? material for backs. For drawers we also use the same type of ?? material that is used for the boxes they are going to go in.

Finished side panels don't have any bearing on box part cutting or assembly since we choose to use applied finished sides. Applied finished sides as compared to integral finished sides are an example of where some get hung up on local thinking. They can't get over cost of using two side panels, one being the box part, the other being the finished or exposed part. While focusing on the small cost of an unfinished box side panel they lose sight of the larger concern - how many completed cabinets and associated trim parts and pieces can be completed per day\week.

The choice of a construction method that requires multiple unnecessary material types causes many areas of waste. At the same time, a construction method that uses ?? box materials with ?? backs is approved by most if not all of the institutions like AWI or WIC.

Wasted time created and lost by a construction method that includes several different material types could have been used to do things that are required to have finished boxes or trim pieces ready for delivery and installation. This time is almost always counted in labor dollars which can be one of the largest blocks of dollars in the cost of finished product. This is especially true when that finished product is a custom cabinet.

A construction method that includes several different material types will most likely increase the size of the work cell, therefore increasing the amount of shop floor space required every day you are in that building. It's the increased cell size that creates the need for carts and labor dollars to roll them around. It also costs more everyday to clean up a larger cell area.

Viewed in the context of the original question about CNC performance, the choice to use a construction method that uses only two material types mean the operator is only spending labor hours doing things that lead to more finished cabinets at the end of each day and this includes keeping the router busy.

Since an applied finished side does not get involved in the process of cutting and assembling boxes, the cost of an extra side panel is very small compared to labor cost required to make an integral finished side show up at the assembly station at the earliest possible moment. This is true if you have either one or two people working in the router\banding\assembly cell.

If we are producing cabinets that we will install, then back material is cut first. An entire job of backs can be stored in very small area right at the assembly area. Because the router is located very close to the assembly area, and the cycle time for backs cut on the router is so fast, the operator can carry a sheet worth of backs from the router to the assembly area while the router is cutting the next sheet. There's also enough cycle time for the operator to sort and organize backs at the assembly area.

In a case where we are cutting box parts out of plywood and finishing is going to be required, I would cut these parts as the first step when a new job is started. These parts are out of different material and are going to go through a different path in the shop compared to all of the pre-finished parts. They will go to an area where they are finish sanded then on to the finish room before they go to the assembly area.

As to the question of if the box we produce is custom even though it uses two material types, the answer is, very custom. Every box we produce is for a different customer. Even in the case where we do a habitat house once or twice a year, these houses are often the exact same house. Sometimes it's mirrored, but very seldom are two of these houses the same. So even in these cases, each box is custom built for one family.

From contributor A:
To the original questioner: What are you using to generate your nests with?

From contributor J:
That machine with a pusher and loading should run that job in a day easy. As for your handling after the bhp 200, this may be your hold up? We run at about the same 30 meters a min at 24000 rpm. Weeke has many customers you can compare your output with. Labeling may speed your machining up. Are you using Woodnest and Office to machine? They may speed your sorting of parts up.

From the original questioner:
I have never checked into a labeler for our router. It may help some. Usually we have the parts marked and carted before the next sheet is cut. As I read other posts on this subject, our output is probably about right, or at least similar to what other shops are getting.