23 Dec Prognostic Analytics vs Predictive Analytics in IoT
t seems like predictive analytics gets all the attention these days but generally speaking, it requires either a Data Scientist or a machine learning algorithm operating on lots of event data, in order to predict the all-important dimension of time, at least to any degree of useful certainty. Enter prognostic analytics. In a closed system of uniform conditions, prognostic analytics can make better predictions about the “when”. In this episode of the IoT Business Show I speak with Moritz von Plate about how the characteristics of this older-school stats make it very well suited for predictive maintenance in Industrial IoT.
In this episode of the IoT Business Show, I speak with Moritz von Plate about how the characteristics of this older-school stats make it very well suited for predictive maintenance in Industrial IoT.
Moritz is the CEO of Cassantec, where he is responsible for strategy, finance, business development and sales. Moritz started his career as a strategy consultant with BCG and was CFO at Solarlite, an award-winning solar-thermal power plant manufacturer.
So what does this mean to the manager? Predictive analytics is applicable to less constrained systems, more open systems, especially ones where human variability is involved. But if your use case is a self contained, closed and uniform system, as is often found in industrial, infrastructure and many commercial IoT applications, prognostic analytics should be considered. Why? Because under these conditions, prognostic analytics can predict, for example, that your part is going to fail in 3 weeks with a 70% likelihood … without the help of a Data Scientist. This is the Holy Grail of predictive maintenance and ironically something that’s not possible with predictive analytics today.
Here’s What We’ll Cover in this Episode
- Which country is the world’s leading producer of surfboards and how is that related to energy generation.
- Why using human Data Scientists to make prognoses on predictive hypotheticals is a problem.
- Why Industrial IoT is a good fit for prognostic analytics.
- Why it’s difficult to build predictive models in manufacturing – it’s all about the sample size.
- Why IoT Platforms are putting OEMs at risk of losing their customer relationships.
- The relationship between Captain Kirk, Mr. Spock and Data… the android.
- The best environments for predictive analytics and prognostic analytics.
Mentioned in this Episode and Other Useful Links
Support this Podcast
If you have been enjoying this podcast, there are a few ways you can support it:
Have an opinion? Join the discussion in our LinkedIn group
What’s your view on prognostic analytics – have you ever heard about it before?Click here if you have an opinion on this podcast or want to see the opinion of others