When I ask a bunch of different sources different questions about different topics, and they all give pretty much the same answers, I start to pay closer attention—mostly to make sure I'm not hallucinating. This doesn't happen very often, but when it does it can be a little unnerving.
Anyway, just over three years ago, I was asked to research the biggest technical innovations in industrial networking for our late, lamented Industrial Networking magazine. Great, another "life, the universe and everything" story. Now, I don't mind covering broad topics, but this one seemed to be far too all-inclusive. Panicking, I asked everyone I could think of about the most important innovations in networking, especially as they applied to process control and automation. If you were contributing to any of Putman's three automation publications at the time, there's I good chance I contacted you.
Well, I learned from the end users, system integrators, suppliers and other experts who chimed in that it wasn't the evolution from hardwiring to fieldbus to Ethernet to wireless that was the biggest innovation in industrial networking. Instead, they all reported that the greatest technical innovation in networking for control was the faster, smaller, cheaper and more powerful microprocessors and all the software they could run to carry out every kind of process control, automation and manufacturing task. Ironically, I even ended up talked to engineers at Applied Materials, who reported their semiconductor manufacturing processes were in turn being aided by the very chips they produced. You can check out this story, "Heart of the Network" in the November 2011 issue of Industrial Networking.
Since then, I've found microprocessors at the root of many innovations in the process control and automation fields. Some I expected, such as the data processing enabling plant-floor systems to get closer to their enterprise-level counterparts in "Shortening Shop-to-Top Trips" in the January 2015 issue of Control. Likewise, less costly and more widely distributed chip and software are finally making distributed process control in the field a practical reality.
What may be less obvious is how much microprocessors are assisting many traditional hardware components and streamlining their configuration, operation and maintenance. For example, when I researched this issue's "Level Best Fit" article on level measurement, several sources almost echoed each other in describing how the essential level technologies haven't changed much recently, but they've been greatly aided by chip-based software and communications, which can coordinate the data gathering of several level instruments, deliver more and better information for analysis, and improve the performance of their applications.
The last time I was this surprised when reporting an article was several years ago, just after I finished several software-related stories. I'd moved on to covering a nice, solid motors topic, inquired what was going on in the field, and then found the main trend was added intelligence and software! D'oh! I was right back where I started in the land of ones and zeroes. These days, it seems like all roads lead back to the chips.
So if there are any technologies that you considered implementing in past years, but didn't because they were too costly or too difficult to program, then it might be a good idea to take a look at them again. Onboard chips or networked data processing and software have likely made them a lot easier to employ lately. I'm not saying any engineers or operators don't know what solutions are available to them, but so many devices are being updated so quickly with microprocessors that it can be hard to keep track of what many formerly familiar devices can do now.
Personally, I would just as soon ignore the breathless hyperventilating over Internet-enabled devices and remote software services. But if a microprocessor is making it easier to deploy and manage components on the plant floor, then that kind of improved performance is frankly a lot more important and useful than making sure the enterprise level gets the numbers it wants.
What would really be cool is if chips in the field and distributed intelligence could be enlisted to carry out even better cybersecurity functions and really give engineers and operators better ways to protect their applications and facilities. I'm sure it's already happening somewhere. I just need to find out where and let you know.