It’s only been in the past several years that the decades-long quest for hierarchical integration of control, execution and enterprise systems has become feasible. Innovative middleware and standardized process models finally are taking some of the headache out of enabling these disparate worlds to work productively together.
I’m not saying there isn’t plenty yet to be done along this vertical dimension of the Purdue model pyramid, but the next priority for manufacturers will be to better integrate the software-based tools—and the people that use them—along an entirely different axis. Namely, through time.
Take the design process for a typical capital project—a new line or perhaps a new plant. In the heavy continuous process industries well-evolved simulation tools exist to aid in process design, but how well do they really integrate with the mechanical design tools they feed? Do process engineers’ systems allow them to “design for constructability,” based on feedback from a plant’s mechanical design systems?
Similarly, sophisticated tools for three-dimensional plant design provide for instrumentation and control wiring and schematics—and in some cases even accommodate control system configuration—but instrumentation, control and optimization strategies still bring up the design caboose.
Now, move away from this best-case scenario in the heavy process industries and into batch/hybrid process industries. Here, the parallel design efforts of multiple machine OEMs and skid builders often must be brought together, and the integration problems are more complex and multidimensional.
In the end, it’s not just a technical integration problem, but a cultural one. Do the people responsible for upstream design functions have the vision to understand and accommodate the priorities of those downstream in the design process—even when a better outcome is ultimately at stake? Do their systems and their mindset help or hinder that end?
In many cases, it’s still over the silo wall and on to the next thing. Our quarter-to-quarter, project-to-project orientation focuses design teams on containing costs for their particular functional part of a project. There’s little incentive for fundamental process and systems changes that might require more investment upfront, but will pay off over the longer term. And, not to paint too dim a picture, this is just the design process we’ve talked about so far. Wait till the plant is built, and we get operations and maintenance’s take on how things really should have been done!
Clearly, the next frontier of systems integration is a more holistic view of design, construction, operations and maintenance processes throughout an asset’s entire life cycle. A more integrated view has the potential to bring production assets on line more quickly and efficiently, and to operate more productively throughout their useful life.
The pieces of a better way already are starting to fall in place, through such concepts as product life cycle management (PLM) and the increased uptake of mechatronics for machine design, a more interdisciplinary approach that brings simulation and automation in from the beginning. (Indeed, the recent acquisition of design software maker UGS by Siemens is another example of the recognized potential for better integration across the across the design-production engineering divide.)
Ultimately, however, full integration across the fourth dimension requires an integrated view of information associated not only with production assets, but with people and processes as well. True process optimization over time will require visibility into and analysis of unified plant data from a number of different perspectives. The genealogy of a particular end product, for example, must include its full batch record—information on the raw materials used, processing temperatures and pressures, the operators and production equipment involved, and even its supply chain experience. Similarly, intelligence tools applied to the people part of the equation are needed to identify operator performance gaps and ensure adherence to best practices. Finally, yet a different set of intelligence tools applied to the asset-centric view of plant information are needed to facilitate the enterprise’s larger asset optimization goals.
I realize you’re not yet done with your sensor-to-boardroom integration tasks—but I wanted to reassure all you Buckaroo Banzais out there that when you’re done, further integration challenges await—across the fourth dimension!