A friend of mine wrote an editorial recently where he predicted the imminent demise of Layer 3– manufacturing execution –of the Purdue Model of manufacturing technology. He hides behind a paywall these days, so I don’t think I can link. Funny thing is, he’s always been focused on the lower layers of technology. For him to try to create controversy here was, to say the least, surprising.
Perhaps a recap is in order at this point. The Purdue Model has withstood the test of time. It described technology and application layers 30 years ago that are still true. Technology is always fluid, but certain things just have to be done in a manufacturing or production enterprise.
Layers 0, 1, and 2 describe the instrumentation, control, and automation layers. Layer 3 describes what has been known as the MES–or execution–layer. Layer 2 describes the enterprise layer–known as the ERP layer.
My writing has focused at the lower layers for the past 18 years. I have some work on the MES and ERP application systems. Prior to 2014 my work was almost exclusively for controls and automation magazines. There remain no magazines devoted to Layer 3. No advertising or promotion dollars exist for that area–or at least not enough to fund that level of journalism. I thought I would focus on that as a one-person digital media site, but there’s just not enough money or news available there. The ERP level magazines have also mostly folded, but there remain huge sites that cover enterprise applications.
So back to the (non)controversy.
Some people have been predicting (hoping?) that connectors could be constructed such that real-time data can flow directly from production/manufacturing to the ERP layer effectively squashing layer 3.
But wait! All those functions performed at that level still need to be done–inventory, work-in-process, scheduling, laboratory integration, routing, and the like. Yes, ERP suppliers such as SAP, IBM, and Oracle wish that their products could absorb the functions of Layer 3 and therefore they could be a one-stop-shop for all manufacturing and enterprise IT functions.
Just as certainly the suppliers of today’s MES solutions–GE, Rockwell Automation, Schneider Electric (Wonderware, et. Al.), and Siemens (plus many more)–hope that that scenario won’t happen. Unless, I suppose, that they could sell their solutions to one of the big ERP suppliers.
The Real Manufacturing Execution Problem
The real problem at this level has little to do with technology or application. It’s the name. MES evolved from the earlier (think 70s) MRP and MRP II. Thanks to the stellar work of the ISA 95 committee, the term MOM has sprung up. And I read more about “operations management” than I do about “execution”.
Operations management holds a clue to the future. It is not all about the technology or the application any longer. It is all about business benefit–to the customer. New technologies such as the rise of importance of analytics and new visualization such as smart phone interfaces are changing the nature of Layer 3. There is still a Layer 3. It may not look like the Layer 3 I implemented in 1978. It may not look like the Layer 3 of five years ago. But the functions are still required, still being accomplished, and getting better all the time.
My friend sometimes tries more to be controversial than enlightening. Controversial gets page views (OK, so I pulled out an SEO headline myself). But I’d rather spark a conversation.
