In the semiconductor world the rate of change is incredibly fast. Processor power continues to increase and costs continue to decline. It makes you wonder if there are any limits to miniaturization. Yet the trend to better, cheaper and faster has continued without pause for nearly 20 years. So it appears to be a safe assumption that things will continue much as they are. Although the notion of feature sizes in the nanometers are quite daunting, computing based on light itself has been maturing in the background, so who’s to say where it will all end up?
For those of us in the industrial controls arena, the forces driving silicon are more difficult to deal with. Product life cycles in the industrial world are generally 5 to 10 years. Manufacturing resources are built on the return of investment. You can’t build a beverage canning plant that puts out a million cans of soda or beer a day without calculating the cost of such a facility and how long it takes to become profitable. It requires an investment of millions of dollars and man-years of experience to make possible a 2 liter bottle of soda or can of beer that costs around $1. So if you are going to do it, it had better last long enough to be profitable or it ain’t gonna happen. The profit margin in the beverage industry is pretty small, so a small profit at high volume is required over longer time frames to justify investment.
The other major problem for the controls industry is that it does not have the volume of usage that a consumer product like a cell phone has. So when you realize that the shiny new phone you are carrying around that you got for $100 (or maybe even for free with your phone service contract) has more computing power than a supercomputer did 10 years ago, you start looking at all the stuff in the industrial controls arena and see that equipment is very expensive and bulky by comparison.
Ethernet, Internet and Wireless technologies have taken quite a few years to become accepted means of interconnection and interoperability in the industrial controls market. This is partly to do with the dozens of proprietary communications standards that have been promoted and with the concerns about reliability of the technologies themselves. As more manufacturers are becoming required to monitor processes and provide historical data, the demand for these types of networks is increasing rapidly.
The other barrier to change is the installed base. The controls industry giants are all billion dollar suppliers. They sell equipment at these rates every year. So the installed base of hundreds-of-billions in control equipment is it’s own barrier to entry. The new stuff has to be a lot better, cheaper and faster than the old stuff, or it may not be worthwhile to replace what you own.
We need new technology to enter the market in an effort to accelerate change, take advantage of declining costs and make possible new applications that were not cost effective a hardware generation ago. A whole new spin on the mechatronic challenge.