AW: What do you feel are the important communication strategies for manufacturers in the 21st century?
Truchard: To set some perspective, what you are seeing with the evolution of off-the-shelf technology is that it keeps getting better and better. If you look in the general marketplace, two interesting areas are telephone over IP and live video over the Internet. Both of these applications require high quality communication service.
In industrial automation, it’s very similar. For a given application, you need the same high level of quality service, with dimensions such as security, reliability and robustness. As the quality and service of commercial technologies become available, people discover them and use them. As they use the new technologies, they move the quality of service to even higher levels.
When the general marketplace drives communication technology to higher speeds—from 10 Mbit to 100 Mbit to gigabit to 10 Gbit—each one of those steps, in a brute force sort of way, raises the bar for quality and service. You can add redundancy and security and still have a very high performance system. That sets the framework from which we’re all working. Obviously, in the end, we’d all like to see our machines, databases, quality, monitoring, scheduling and inventory management systems tied together. For example, the quality and service of Ethernet components, such as switches and routers, have greatly improved over time and provide the opportunity to create private networks. These are all tools to integrate equipment and make the communication system work.
AW: What are some of the technical challenges manufacturers face in using new communication technologies?
Truchard: Technically, it’s two levels—at the enterprise level and the manufacturing level. I’ve been working at National Instruments on our next generation of tools that serve the enterprise level for high-performance automation applications. Around here we talk about first-class distributed LabView software that has the ability to tie things together in a variety of ways for very high performance in real-time applications. Some of the specific technologies we are looking at include global clocks, publish-and-subscribe and very tightly coupled systems that use high-performance network schemes such as FireWire or USB [universal serial bus].
Each one of these technologies opens up a new opportunity, a new way to build your automation system so that it works interchangeably and interoperates to tie different kinds of capabilities and equipment together in the same manufacturing environment.
AW: What are the cultural challenges in adopting new communication technologies?
Truchard: Historically, what automation vendors did in our space was fairly isolated from the enterprise and information technology space. Now we embrace the same technologies as off-the-shelf IT products. The IT department is getting more involved. As well, the need for connectivity to the enterprise-level applications drives IT involvement. If you want to move data from the manufacturing machines into quality and production systems, you’ll find yourself more and more working with the IT department.
AW: How can manufacturers overcome these challenges?
Truchard: Here’s a specific example. Automotive manufacturers want enterprise-wide collaboration in the design process. One of our products, DIAdem, which was part of an acquisition for NI, works alongside LabView for post-acquisition data analysis and report generation. With DIAdem, hundreds of automotive engineers can share design and production data across the network in an accessible database, for instance, in order to pull up data for test runs and crash tests. This is an example of enterprise-wide reuse of data for collaborative analysis in the manufacturing and design cycle.
AW: What drives a manufacturer to improve its communication strategy and bring everyone involved with the manufacturing product and process into the communication loop?
Truchard: First off, it’s efficiency. Perhaps quality data must go to the design engineer. Unless you have systematic ways to do that, it becomes an ad hoc procedure and not repeatable over the long term. Automating the way data are shared ensures that it can be routinely done and not done just as a firefighting effort.
Visionaries in the organization may take action based on intuition. An engineer may say, “Gee, I know I need this data. I know I need to connect things together.” They hear about a technology and decide to apply it because it can help them do a better job.
There’s a lot of pressure on costs and doing things more efficiently. These can be drivers as well. For instance, I was reading how in the drug manufacturing industry, when a drug is under a patent, no one is too concerned about efficiencies. As it gets closer to the time when the patent will expire, they suddenly get very excited about efficiency.
AW: With networked systems, vast amounts of data are now available. How are manufacturers deciding which data they really need to pay attention to?
Truchard: This is where post-data acquisition products come in. Manufacturers first have to efficiently store the data and then know where to go to find it. They need to create a tool that lets them find where things are very quickly. It’s kind of like the Internet itself and using a search engine such as Google. You can find virtually anything you want in seconds.
There’s an incredible amount of information available, but you need a really efficient way of finding it. It doesn’t matter that there’s lots of data, as long as you have an effective way to get to just what you want. Archiving the data so it’s readily accessible the way an engineer wants it is, in our view, the key to being successful. Being able to find a particular piece of data should be the focus more so than deciding which data to keep. Storage is inexpensive. Creating good ways to find what someone is looking for is the most important task in a communication strategy.
AW: What do you say to those who aren’t even sure what it is they should be looking for?
Truchard: They must come up with good ways to label and tag things automatically so when they figure out what they’re looking for, they can find it. Communications technology makes it possible to do that. It’s much easier to automate in difficult environments, such as an offshore drilling platform, than to put people in those expensive and dangerous situations. You can minimize personnel hazards by having onboard sensors, automation equipment and communication technologies to get the data to the right place. We just acquired a product called MATRIXx [simulation and control software] that automates various functions on the space shuttle—that’s about as remote as you can get!
AW: How do wireless communications technologies, and their inherent concerns about reliability and security, play a role in the control network?
Truchard: There are two aspects. One is replacing fixed wiring, which is an economic decision and entails looking at performance requirements. The second is the “move around” use of wireless to carry information along with the product. For instance, LabView software can now run on a PDA [personal digital assistant], so you can walk up to a device and ask, “How’re you doing today?” Inspectors or maintenance personnel do not have to go back to a terminal. They can operate by wireless.
AW: How important is it to have one communication standard for manufacturing networks?
Truchard: Well, I really doubt that we’re going to get to one. However, it’s kind of the perspective from which I started. That is, that telephone will ultimately work over IP and you’ll be able to get real-time data over Internet or cable, for example. What that’s really saying is that, in time, it won’t matter. The network backbone capability will be so good that it will be like your telephone—it just works. The customer will just point to whatever technology he has and say, “Make it work.” In the meantime, we have to wrestle with each of these proprietary systems and standards. Our goal is to work with and support as many of these systems as we can and make a decision based on open technology.
The off-the-shelf commercial solutions sell at a fraction of the cost of proprietary networks and systems. Everyone will have to play on this open, high-performance backbone.
AW: Across the board in manufacturing, once something is working it’s difficult to implement a change unless there are some pretty powerful drivers and benefits. And then, once a system is in place, it may have to be supported for 20 years. How do you reconcile the need to adopt leading-edge technology while at the same time support your automation system through its life cycle?
Truchard: That’s a question that has come into play with the use of personal computers. At National Instruments, we try to add a layer of technology that insulates the customer from this rapidly changing infrastructure. I’m sure that many of the other vendors view this the same way.
Manufacturers can’t adopt every new standard and capability that comes along. You want to make sure it has a reasonably long life cycle to minimize future support problems. Our APIs [application programming interfaces] have been supported for almost 20 years. National Instruments puts a lot of energy and effort into creating baseline technology that can adapt to change, and not become obsolete.
What you really want the customer to be able to do is keep things the same at the top level in the layers that cost the most money, such as in software, but change out the infrastructure. This is perhaps the key trend—that we as vendors will do a better job at allowing our customers to update to the latest infrastructure technology while maintaining the customer’s investment in software and top-level system design.
See sidebar to this article: Eating your own dog food
See sidebar to this article: Dr. James Truchard profile
