Everyone is talking about the internet of things and how this will change how we interact with the world. Having what is essentially a connected computer monitoring your operations opens up a lot of opportunity. Not only for delivering data to be analyzed, but additional functional capabilities for the business and for the devices.
Bring connected gives the manufacture (or others) the capability to correct and enhance their devices at any time. For example there is a company called Sonos www.sonos.com who make a wireless audio products including speakers. Recently they have added capabilities to their solution, without requiring the purchase of additional hardware. By updating the software capabilities of their speakers and adding additional capabilities to their controlling software they added “trueplay” to their solution. This gives the homeowner the capability using an iPad or iPhone to fine tune & adjust their Sonos speaker to the room that the speaker is in. Great capability and this was not part of the original installation. In the spirit of full disclosure I have been a satisfied Sonos customer for many years.
So off I go to tune my speakers. This for me is not a large task, just one that I have to learn how to perform, schedule (due to the requirements e.g. minimum background noise), and perform. And then sit back and enjoy. A good example of what the IoT can deliver, new functionality delivered with no extra cost.
But scaling this process up to a manufacturing facility adds additional complexity. It is quite common in the software business for the vendor to release upgrades and patches. But as everyone know not everyone applies these releases immediately or ever. Software normally is installed on a test system, tested within the landscape for compatibility, and a selection (maybe all) of the processes tested, before the software is installed in production. A normal part of the IT process and good IT departments will have anticipated the need and budgeted accordingly. However IT has this luxury, but does manufacturing? Or should it?
In the manufacturing facilities that I have been involved in, none of them had redundant manufacturing control systems, which we could “play with”. So what happens, in the new IoT environment?
It is my belief that manufacturing and operations will have to start applying some of the IT principles. We will have to include a test environment in our facilities so that we can isolate the changes and test these changes properly. I also believe that IoT will drive more standardization on the shop floor. As it becomes more important to keep the various equipment software releases up-to-date, it will become obvious, that just as IT which only supports a limited types of equipment, manufacturing will only have the resources to be able to test, and install / support a limited number of types of equipment.
This limitation consequently will drive a more heterogeneous shop floor environment. Just as with IT, manufacturing will have to reduce the number of variables that needs to be accommodated. One way of doing this will be by reducing the different types (manufacturers) of equipment on shop floor. This pushes some of the integration responsibility back onto the manufacturers. Those manufactures will have to guarantee that the upgrade will work with all their equipment. This leaves manufacturing just to test the integration points to other manufacturing and IT systems.
While IT has the luxury of being able to upgrade their systems at any time (especially the PC’s) manufacturing does not. Unless there has been a problem, any upgrades / changes to the shop floor are generally down during a shutdown / turnaround / overhaul (STO) process. This just adds another process to the already crowded STO schedule, but is still doable. Whether operations does the upgrade themselves or subcontracts the work out, it still falls into the standard STO planning and execution processes. The area that is new and still needs to be thought about, is what happens if the upgrade delivers a new processes (e.g. the above SONOS example).
I believe that including the investigation, testing, and implementation of these new processes would be too much to be included in a STO process. After all the main objective of STO is to get the facility back into operation as fast and safety as possible. So adding additional (and potentially non value add) processes into the schedule does not make sense.
This then brings me back to the idea that manufacturing will have to invest in a test bed area so that the new processes can be investigated. And, if of value, then these new processes can be scheduled to be implemented, and if necessary, can be incorporated into the planned shutdown.
IoT will change how we run our facility’s and will offer a lot of benefit, but nothing is for free. IoT will impact a lot of processes that are not at first glance tied to IoT. How we handle them will have a significant impact on the success or failure of the whole IoT initiative.