If you want to supply machines to the big boys of consumer product manufacturing, you better smarten up. Smarten up your machines, that is.
That message came out loud and clear at Rockwell’s Automation Fair, which this year hosted a forum for OEM machine builders. Speaking before a crowd of a couple hundred machine builders, engineers from Procter & Gamble and 3M talked about their growing need for “information enabled” production and packaging machines.
Whether they’re turning out beauty products or tape or putting product into packages, machines used by the consumer products companies still have to meet stringent functional requirements. “The machine has to work properly. That’s always a given,” said Ron Burger, a systems engineering specialist with 3M’s Corporate Process Instrumentation and Control Systems Group.
Yet working properly isn’t enough nowadays. As savvy end-users optimize their production processes-–though practices such as measuring Overall Equipment Effectiveness (OEE)–they increasingly require smarter machines that can talk to one another and to corporate business systems. This requirement for smarter machines has all kinds of machine and control design implications.
Take something as seemingly straightforward as control panel real estate. Burger explained how 3M’s growing desire for process data, advanced monitoring capabilities and flexibility to accommodate process improvements can be seen in the panel design. “I’ll go to an OEM and see a beautiful, full control panel, and I have to ask, ‘Where do I put all my stuff,’” he said, adding that 3M requires its OEMs to leave extra panel space for this very reason.
The same thinking applies to controllers. Burger reported that 3M engineers routinely will specify souped-up PLCs whose processing power might seem like overkill for small-machine applications. But “big PLCs” offer more integration opportunities and can accommodate unexpected I/O down the road. “So we’ll ask the OEM to install flex I/O drops and we’ll provide our own PLC,” he said, arguing that this approach gives 3M the flexibility it need as its manufacturing processes evolve.
Over at Procter & Gamble a similar push toward information enabled machines has emerged over the past few years. Mike Lamping, one of the company’s engineering technology leaders, said that P&G’s ability to improve efficiency in the short term and to make continuous process improvements over the long haul come down to moving the right information to the right place at the right time.
Yet Lamping drew a sharp distinction between raw data, which can often be pulled off individual machines with ease, and the information needed to drive optimization efforts. “Data does not equal information,” he said. “Data produces information as it flows up and down through the business systems.”
“Controlling that flow has been a major challenge,” he continued. But it’s a challenge that P&G engineers have overcome by relying on a standardized approach to automation that includes the use of Ethernet communications, standard tag names for data and a common look-and-feel for HMIs.
This standardization is also reflected P&G’s machine control systems, which increasingly make use of state-based control model spelled out by PackML. By embracing a standard state model that cuts across different types of machines, as well as similar machines from different OEMs, P&G not only improves data consistency but also cuts the training, commissioning and maintenance costs that stack up when you deploy too many one-of-a-kind machines.
Lamping likened the experience of using standardized information-enabled machinery to driving a car. Whether you’re driving a Ford or a Chevy, the basic operation of the vehicle doesn’t change all that much. “If you can drive one car, you can drive another with a few minor adjustments,” he said. “Why shouldn’t we have the same thing for our machines.”