Industry 4.0 is all about the wise capture, analytics, and use of data. Typically companies seek to improve the speed and quality of decision-making, often automating the implementation of rules-based decisions, in their IIoT efforts.
Digital transformation uses data for those, and even more forward looking scenarios that currently are impossible or very time consuming to consider.
As always, the first and most important question is “why?” Why do we want to begin collecting and using more data? Again, better and faster decision making is often the answer.
Actually implementing decisions automatically requires that various data sources can connect with one another and initiate action of some kind. Providing information to improve product capabilities or reliability is a different set of data and different processes.
The what, when, and where of data collection depends on the problem you are addressing.
Your steps into Industry 4.0 will require comfort with data analytics at some level. The basics you can likely handle with existing staff even at a smaller company, but without an intro to statistics level of understanding even that can be dicey.
Yes, algorithms exist that can do the math, but who will determine which algorithm is appropriate? Your engineers will have some basic statistical understanding, but remember they study math much more than statistics.
To begin, first answer the business question of why? Once the high-level why is understood, prioritize based on low level potential as you learn. And again, just because you can doesn’t mean you should.
For example, collecting data on whether or not a machine is running is easy, but who would do what when based on that data? Correlation of data is often important. Measuring the temperature variation of the fluids in a machine can be useful, but is more useful when correlated with production data, machine speed, or other variables that help you understand if the temperature variation matters.
Every manufacturer has long been collecting data, much of it of poor quality and useless. You can’t afford to continue accepting those weaknesses. As you begin a pilot, draft data governance rules and responsibilities. Just as you’ll learn from the pilot, you’ll learn from the draft of data governance. And remember: More is NOT necessarily better.
Once you’ve begun a pilot, enforce assessment, learning, and then move on. Improve that pilot, or better yet, expand that experiment.
Pilot purgatory is no better than ignoring Industry 4.0 completely.
Podcast: Download