Manufacturing can be a simple business and quickly become an extremely complex one. For more than a decade, I've attended automation trade shows, usually wearing my newsboy hat. Every year, I leave with the same observation: many solutions are presented as though every manufacturer operates the same way.

One facility may have a team of controls engineers, while another relies on a maintenance manager who wears every hat from electrical to facilities.

I have worked in facilities and have worked for clients that fit both modes. A common theme I see is the options are plenty. And industry solutions providers are glad to sell their single-source solution. For example, I was recently at a show in Chicago this summer and stopped at a booth for a robotics company. They would provide a one-stop, full system along with service after the sale. I was hooked. Then I asked, “Where does your solution start in the client’s system?”

I received a puzzled look in the salesman. He said, “It is a complete solution.” I then pressed on, using an example of a dog-food packing facility where the complete solution would move from the raw goods to a kibble, to filling the bags with kibble, and then stacking the bags on pallets. This process is simplified for brevity, of course.

From that description, the one-time “complete solution” went from the exit of the machine filling of the bag to the stacking on the pallet. So then, I pointed out, a maintenance team would need to have a new system to master from the filling machine, which included a robot, new processor, different conveyors, different motors, different motor controls and so on.

I tell this story for a reason; I liked the solution and think it is a good choice. But it is not a solution for everyone. I would even say there is a very small market that this is a great fit for. It is an OK solution for many.

I use it as an example of how a cool solution to the purchasing team becomes a potentially large added task to the maintenance team. And, in my experience, the maintenance team is last to be invited to the table of discussions, and, when they are there, they are not as equipped with the current technology available until they get it into practice. By then, it is already installed and needs to work, even if it was not the correct tool. An example would be using a sledgehammer for a 16-penny nail. Does it work? Yes. Is it not only less than ideal, but also harder than it needs to be? Yes.

This philosophy has spread by both established automation component suppliers and new “smart” solutions with venture-backed money startups that are providing a single solution for making an Industry 4.0 solution.

I have seen suppliers connect their energy-monitoring equipment to motors and other large loads to tell you where you are spending money. “Just use our online system to monitor your data. Subscribe with a monthly fee.”

Get your subscription to Control Design’s daily newsletter.

Another trend I have seen is based on failure predictions. Add my module that measures vibration and temperature to your machine, send your data to my online portal, and one more monthly fee. I know a facility that did this, and the maintenance director had a monthly meeting with the “customer service manager (CSM),” where they kept adding tasks to his PM list for a mixer that had a harmonic vibration at 200 rpm speed. The mixer operated at 800-1,000rpm. It only had that “excessive” vibration for 2-5 minutes on startup. It had been torqued as much as it could; the thing just did not like 200 rpm. But the CSM would not remove it from the PM task list because the “smart” cloud would flag it. I’m not even sure the CSM could remove it. So, every week, a maintenance worker had to touch that mixer and answer why it was in failure mode, even when it was not. And they paid a service fee for that extra time that was spent on that task.

The subscription service model of clouds and software-as-a-service (SaaS) products feel like streaming services to me. We cut the cable to get on-demand shows for lower cost, and now we have five subscriptions where we still get to watch TV, but it’s not the solution we thought we were getting, nor did it really save money.

Manufacturers, as I see it, need to own their architecture—one built on the standard technology solutions that they are comfortable with maintaining. Know how it works with their process and why it works for them. Then find new adaptations to apply to that process and technology. Sometimes a subscription is the right answer, and sometimes a task of packaging off loaded to a single solution provider is the right answer. But do it with a purpose and coordinated with the team involved with using it, maintaining it and servicing it.

Common architecture, owned by the manufacturer, is a key in making sure system integrators or aaS solutions are harmonious with the systems. The most successful manufacturers I have worked with are not the ones with the newest software or the most impressive dashboards. They are the ones that have built a common architecture where machines, sensors and business systems share information through open standards. They can replace an application without rebuilding the factory because the factory owns the data, not the application.

When a solution is presented, do not ask what the software can do; ask how it fits in the current system. Who owns and has access to the data, and how is it allowed to be used? A smart factory is not defined by cloud subscriptions. It is defined by the ease of access to adapting to change as it comes along.