What is really new about the Industry 4.0 concept is the paradigm shift in which the collected data is used to make decisions with automated systems. This is the only way to generate added value in the form of higher quality, flexibility or efficiency from the huge amount of data collected. This becomes clearer when we consider how good quality decisions can be made. The methods can be categorised as follows (with increasing complexity):
Descriptive analytics: This means that data is processed and people use it to make decisions. Tools for this are characteristic values (mean value, standard deviations, etc.) and suitable graphics. Excel is typically used as a tool.
Predictive analytics: These methods have been developed since the 1940s and have become increasingly important with the advent of computers. Statistical procedures, machine learning methods, neural networks and others make it possible to make the right decisions based on a small number of rules. The decisions are typically made from three to six alternatives. The decision-making process is therefore more ‘objective’.
Prescriptive analytics: Only with the ability to measure large quantities of data and process it using modern numerical methods is it possible to develop such decision-making systems. Within the Industry 4.0 paradigm, they are the innovation that creates new opportunities for value creation. The systems are able to make optimal decisions from a very large number of possibilities. They can find optimal solutions where a human has no chance of gaining an overview of the process. Ultimately, the basis for such systems is mathematics.
A typical system for predictive analytics consists of the following components:
Mathematical models that can depict reality
Numerical methods to calculate these models
Data collection and processing using methods from signal analysis and statistics
An automatic decision-making system that finds the optimal solutions from millions of possibilities.
Last updated: 01.04.2025
