I am a venture capitalist, one of the early builders of the programmable controller, and a Harley guy. [Editor's note: Morley is considered the father of the PLC.] Enough about me. Let's talk about the new robotics.
The new robotics use machine-to-machine (M2M) concepts. Most corporate assets are not very well monitored; some large companies monitor only 30% of assets in real time. Utilization of unmonitored assets is usually low. For example, if you look at machine tools, process inventory, and human resources, you may find that under-utilization is rampant.
We need a mechanism that can connect all these machines and assets in a network compatible with modern database and Internet/Web technology.
Until now, most Web and Internet activity was people-to-people. There's a change acoming. We want to know the wellness of equipment, and the rate of return generated by net assets employed across the corporation, and we want the means of gathering that information to be M2M.
Machines need not have a human involved in most of their maintenance, service, and status issues.
Let's use a parable here. My wife's kitchen has one occasional cook and two or three occasional eaters. But, the types of electrically powered equipment (and devices that are not electrically powered) clearly number ten times more than the humans.
When walking down aisles in factories, we see many machine tools and not many people. The object of M2M is to connect, network, coordinate, respond, and transmit data to the Internet on a wireless basis. Installation should be simple and straightforward.
Imagine the bulldozer of the future. It's working in open-pit mining. Most of these bulldozers are owned by third parties, not by the operators or mining company. Inside this bulldozer, we imagine a wireless device (computer, memory, and radio) that monitors the bulldozer's characteristics. In doing so, it looks at vibration, oil temp, fluid levels—the entire dashboard.
If a fault arises, this monitoring system transmits the fault status over the Web to a logging depot or server, and also estimates when the service will be required and why. It also orders any needed part. So, if the bulldozer feels ill, it schedules its appointment with the doctor, and notifies the pharmacy as to what medications are needed.
If we think about it a little more, all machine tools and, as a matter of fact, all enterprises, are merely peripherals for the computer. Everything in the future will be connected wirelessly to the USB port.
Ah, but I didn't say much about the human. The human is required in cer tain areas: installation, overview strategy, and design. Many modern technology systems use the statement "all I have to do is..." This is clearly bad. There is a new cell phone called Jitterbug that is only a phone. It does not have a video camera, or download music. It's a wonderful device, and is touted as one of the significant new products of 2006. Its ease of use is great. We need devices for the human that snap into place and are extremely easy to set up—a screwdriver and some wire should be all that is necessary for installation.
A simple design is not unsophisticated. It's the hardest to design. Think about writing. It's easy to write something long, but it takes a long time to write something that's short and to the point.
The technical design of our M2M system will be wireless, and include a self-fulfilling language. The hardware should be readily available and treated as a system component. We have played here at my barn studio with some suggestions about language. We've found that the average person can program a demonstration using a model language without an instruction manual.
The contractor who does system installation for off-road vehicles, factory machine tools, CNC machines, inspection processes, automatic transport, materials handling, and building control will use M2M capability with standard units. The design lets the contractor bid and install systems with profit without too much customization.
What do we do in the near term? Well, contractors are designing servers that have software for such monitoring. Openpit mining, machine tool coordination, optimization of assets, automatic scheduling, maintenance requirements, spare parts, and inventory reductions are all aims of an M2M component system.
The robot of the future is, indeed, your enterprise. It is a computer peripheral that will run by remote control and be able to organize itself.
My prediction? Less than 10 years.
This article was first published in the April 2007 edition of Manufacturing Engineering magazine.