Billion Dollar Business

CEA: How long have you have been with Rockwell Automation?

Deken: I came in via the Reliance Electric acquisition in 1995, as Director, Product Marketing, for the Standard Drives Business. At Reliance, which I joined back in 1974, I had a number sales and marketing roles, including being marketing manager for drives and PLCs. And I have been in my present position since 1999. In that time the Logix controls business has gone from $28 million to almost $1 billion.

CEA: The Logix control platform has been around for 10 years now?

Deken: Yes, almost, the first product was released in 1998. Development work actually started in 1994, at a time when computer technology was beginning to have an impact on PLC technology; the idea being to take the communications capability and programming flexibility of PCs and apply to the PLC world.

Customers were also telling us that it was expensive to have different types of control architectures on the plant floor, for discrete, for motion, for safety, etc. For the first couple of years the Logix functionality was kind of light, but by 2001 it was up to a point where we could offer a clear alternative for customers to switch over and we were picking up clients outside of our normal area, particularly in motion, through the Integrated Architecture solution that Logix provides.

CEA: So should we expect Logix to be replaced soon?

Deken: You know, people do say, “you had the PLC-3, the PLC-5, ControlLogix, so what’s next?” Well, the Logix platform was not designed to have a sea change come after it. But what we will continue to do is develop it, modernize it, and extend its capabilities for use in a broad range of applications.

For example, update the code from C++ to Java; move to a .NET structure to support the Internet better; have programming and configuration packages that are Web based so that the engineer doesn’t need to have the latest version locally; use faster, multi-core processors to deliver faster capabilities and higher performance; and look for more applications in industries like process.

So basically, there is no advantage in throwing away the code and starting all over again. What Logix provides is, essentially, a factory operating system, and it is the most modern control architecture in the world. To go from PLC-5 to Logix was extremely expensive – we spent almost half a billion dollars developing it, and I have a team of about 350 on the hardware and firmware side and another 400 on software.

CEA: We’ve been hearing about “top floor to shop floor” integration for some time now. Is it happening?

Deken: The IT area and the plant area are not really linked up well yet. Where it is happening, it tends to be more of a “forced march” – the interfaces are expensive and do not provide full service capability. The goal is ad-hoc data access – where you don’t need specially engineered interfaces to get timely plant floor data. And that’s where Ethernet comes in.

We use the term “river of Ethernet” to describe a future where the Ethernet will be like a utility in the factory in the same way as power and water are today, and everything simply plugs into this single factory network. For example, all our drives have EtherNet/IP built in and it’s being added it to all of our I/O adaptors and subsystems. We’re also working closely with Cisco, bringing IT and plant guys together to discuss what needs to be done.

CEA: The ODVA CIP networks also fall within your area of responsibility?

Deken: Yes, the development and marketing of the EtherNet/IP, ControlNet, and DeviceNet networks is done in my group. EtherNet/IP, in particular, is really taking off. It’s the most open network available (we gave the rights to ODVA) and uses standard, unmodified Ethernet, unlike some other offerings.

And Schneider Electric recently deciding to standardize on it means that, plus ourselves and Omron, we have three out of the top five factory automation players in the EtherNet/IP camp, and just Mitsubishi Electric and Siemens outside. So it’s a great choice – for discrete control, safety, motion control, real-time I/O.

CEA: Does the prevalence of Ethernet mean less emphasis on DeviceNet?

Deken: At the device level, you have AS-i bus, which has very low node cost but also extremely low capability, and you have DeviceNet, which has higher node cost but also higher capability. And now you have the new ODVA CompoNet network for bit-level devices, which is very interesting.

What’s still unclear is how far down Ethernet will dive down. If it’s deep then that makes a network like CompoNet more attractive. If not, then DeviceNet will continue to thrive.

My guess is that what we’re probably going to end up with is Ethernet for everything except the low-level devices, which will use a simpler, low cost network.

CEA: How about Logix for the process industry?

Deken: Absolutely; we can go head to head with the DCS players. In fact, we have a number of advantages: being able to handle the discrete side in hybrid applications; to talk to many more networks; and to connect a variety of devices from instruments to drives to vision.

Of course, what we don’t have is the 50+ years of heritage in process that companies like ABB, Honeywell, Yokogawa have. But we’re working hard to get that applications capability, and we have added more than 100 apps and support people dedicated just for process.

So I’m here to say that whenever we have a chance to compete against the DCS guys we can win, especially if the customer is able to understand the benefits of having a single controller across a wide range of plant.

CEA: Is it more of a mindset issue then, of Rockwell not being traditionally associated with process?

Deken: To a certain extent. I mean, we have enjoyed the same advantage on the discrete side for many years and built tight relationships with our customers, so we know how difficult it is for other companies to get into our customers. We need to be in a position where if the process customer gets into trouble with an application, he is confident of getting the same help from Rockwell that he feels he can get from the likes of Honeywell and Yokogawa.

In addition, the discrete side is still sold primarily as hardware through integrators, the process side is sold directly as a solution – the client doesn’t care so much about the product but more that the requirements are met. That means that we almost have to have “two faces” to the company.

But our ambitions in the process sector are serious – along with internal expansion, we have the alliance with Endress+Hauser, and we recently acquired ICS Triplex to get triple modular redundancy capability for process safety. So we’re knocking at the door of the Big Six and doing our best to break through.

CEA: Finally, what do you see as the end game for Logix?

Deken: The world is headed towards a two-tier plant automation architecture. The PCs out on the factory floor today are going to disappear – the operating system, hardware and I/O standards issues are insurmountable problems.

Instead, things are going to go the other way, with PLCs becoming more powerful and becoming communications hubs that link to the business systems. The other pieces are discrete, safety, motion, process applications all using a common architecture, and with IT and the plant all tied on Ethernet.

So that’s where we’re trying to go – Logix controllers everywhere acting as the plant operating system and serving data to higher level systems. It’s an achievable vision.

The Logix platform