There are a number of reasons for pharm. manufacturing to resist IoT. It' s ironic that a 'Quality' Organization performs the study.
A. It needs be applied to a large I/O application; no point in putting the energy into a small system. So, the entire large system needs validation of the (over-laying?), or intimate SW and applied algorithms. Bottom line: It's NOT done in-house; so from the get-go, there is a 'quality', reliance on an outside third party. Failure at 4 in the morning- call them. Need to upgrade the SW,: call them. Need to replace worn tired devices that (may) or maynot need updating: rely on THEM. You see where the costs, potential down-time, and sustainable 'quality' may suffer.
B. 50% of the time, the errors in a system are not the working system, they are failures in the sensors; adding MORE sensors can increase that ratio.
C. FDA rules regarding any device labelled a 'Processor' requires special attention if that part needs replacing. I.E. : if a processor needs replacing, than the replacement requires TOTAL revalidaton. It must have the same REV, same, prom rev, same everything... and then requires revalidation. The goal in keeping a system, or device serviceable is to keep PROCESSORS to a minimum. As a simple example: if you have a cold room whose electronic thermostat fails, you relpace it, and reprogram it's settings. IF that same device was 'described' as a PROCESSOR in the system description in stead of calling it a thermostat, then you must, by FDA guidelines, re-validate ALL the functions (bracket the temperature use range) of that cold room again... which means empty it, and spend three days documenting performance. Avoiding 'smart processors' and minimizing periphreal 'processor' based devices that could negatively impact the 'quality' of a product, impact inter-process down-time, or lengthen down-time is to be avoided.
While it might be more convenient to use an internet connection to monitor each machine in amplant, it is not nessesary to do it that way. Tenty five years ago the company that I worked for provided direct communication systems to some of their machines installed at a major auto plant, with no internet anything anywhere around. The best part was that the connection was quite secure from end to end, and nobody in the IT group had to do anything to protect it. It is still possible to have total data reporting using hardwired connections, and know that a system is safe from outside hackers. While the initial expense of installing a wired non-internet connection would be hired, a whole lot of money would be saved by not having to guard against the thousands of hackers on the internet. The fact is that there is indeed an advantage to having a system be totally private and speaking a language unknown to outsiders.
Of course it would make providing access to everybody a lot less convenient, but is having an intruder-proof system woth that inconvenience? For many companies it would be.
Rich, at our house we are approaching the number you mentioned very quickly. Actually, one of ours moved away, so we experienced a drop in the number of devices. That is back on the rise, though.
I would expect on the smart manufacturing end, the issue is sunk cost. Most industrial machinery needs to be kept for a period of time to justify the cost. The 13% penetration you mention says to me that there is lots of room to grow in this market.
Earlier this year paralyzed IndyCar drive Sam Schmidt did the seemingly impossible -- opening the qualifying rounds at Indy by driving a modified Corvette C7 Stingray around the Indianapolis Motor Speedway.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.