I’ve started to work on yet another batch of Edison lamps… I decided to use the principles of my education a bit this time. I’ve been talking about cost allocation approaches and ways to optimize output over the past couple of weeks in my big boy job. The basic idea is that there are certain factors that drive cost or dictate yield. It actually means much more than that, but as applied to my one man shop, that’s the basic idea. For the lamp builds, the concept is that in addition to raw material costs, I bear a cost (in terms of time) to build each lamp; and there should be some optimal number to build that minimizes my average cost per lamp while producing the most high quality lamps.

The way to optimize my time is to understand what are the drivers of my cost. For example, every lamp build has common steps including the first steps of dimensioning the lumber. In order to do this I must joint a face and edge of each board. I then resaw one board for the top. Finally I plane all boards to final thickness. There isn’t much time to save here though, because the time to setup each piece of equipment is minimal. There is zero effort on the jointer, because there isn’t any “setup” that is required. As a result, the time varies directly with the amount of lumber I’m jointing. This discussion isn’t all that interesting, but the point is that if I’m building 1 or 30 lamps, I can’t save any time by building more than 1 lamp at a time. Once this step is completed, I ended up with the stack of lumber seen in the photo below.

Once all of the lumber is jointed and planed, the next step is to get the boards to final length and width. In this case, measuring is required. As a result, there is effort to position the table saw fence to get the boards to the right width. Similarly, all lamps are 4 1/2 inches deep. This is another case where I can setup the equipment once and just cut as many side pieces as quickly as possible (while being safe of course). Technically, I could cut the sides for 1000 lamps at once and save significant time. But if I took that approach it might take months before I finished a lamp. That’s both bad for customers waiting on lamps, and I would be carrying far too much inventory cost. A little time is saved here, but it isn’t super significant. Once that step is finished, I ended up with the stack of lumber below.

Similar to the last step, cutting the dovetails can generate some real cost savings. There are a lot of setup steps (setting the depth of the router bits, changing the router bits, and setting up the jig). Each of these steps takes considerable time. This is one of the major areas where I can save time. Once all of the routing is finished, I ended up with the pieces in the photo below.

Ultimately, I could treat this as a big equation. I could optimize the build quantity for the time and effort to build, controlling for inventory costs, changing raw material costs, travel time to the shop, etc. But ultimately, I’m building these lamps because I like spending time in the shop working with wood. The reality is that I decided to build a few lamps this time, because I want to have some in inventory to sell (as opposed to building them all to order).