I recently put a bow on “Season One” of my Industrial AI podcast series over at the This Week in Machine Learning & AI podcast.
The series coincided with the development of my industrial AI report, which I’m very excited about. Why? In part because of its back-story: what was originally intended to be a short overview of the space turned into a 28-page long, extensively documented (36 references!) paper by the time I was done. But I also think it does a good job of defining the opportunity around industrial AI, categorizing and exploring industrial AI use cases, and exploring various challenges and solutions.
The report develops the idea of monitoring, optimization and control as a taxonomy of sorts for not only industrial AI problems, but for AI-enabled business applications as well.
In fact, I’ve found these to form a maturity model for AI adoption across various use cases: Most companies start by using AI to help them keep an eye on things, as they grow in confidence they use it to make those things better, but with manual intervention, and when trust is established, they give AI the reigns and allow it to exercise control.