According to many professional journals and experts, big data is the key to success in predictive maintenance. Apparently, a lot of information is needed to be able to predict when an asset or system will fail. Experts agree that a lot of data is needed for this type of maintenance. It is therefore important to look at historical data availability, suitability of information that is necessary for implementing predictive maintenance. The credo apparently applies here: the more data, the better. Although many people talk about information, it is unclear what it is exactly. This is related to the fact that information is used for various purposes. The definition (or the lack thereof) has an analogy with the word ‘energy’ in the 17th century. We see it as an important phenomenon, just as it was then, with different appearances, but we do not yet know exactly how to talk about it.
From a semiotic perspective, information can be seen as a series of signs or signals that can be transferred from one (physical) system to another. To be able to influence the behavior of the receiving system (pragmatics), or to derive a specific meaning from it (semantics). Based on the way in which systems and assets deal with information, pragmatic and semantic systems can be divided in a pragmatic system electronic symbols and / or signals are arranged mechanically. This type of syntactic programmed systems (CNC machines, process installations, lock complexes i.e.) cannot interact with the environment and do not know it either. In itself there is no pragmatics or semantics in the interaction with the environment. Research into AI, Machine Learning in the past focused primarily on this type of ‘automata’. What is missing is a mental representation of the environment in which the system must function. What is needed is a “brain” as an intelligent part of a larger whole. This is not the same as a pre-programmed interaction!
More intelligent, on the other hand, systems do contain a syntax. Rules are contained in rules of conduct and feedback, on the basis of which information is exchanged with the environment. These complex systems are semantic in nature. The interaction takes place from the conscious experience of the environment. This type of assets does not (yet) exist. From the point of view of maintenance, this offers possibilities (and only then) to indicate the time of necessary maintenance itself. This as a reaction too, for example, the harsher environmental conditions, predicted use in relation to the current measured wear, remaining life in the planned Life Cycle and so on. The intelligence required for this presupposes the presence of goal-oriented action, rational decision-making and effective interaction.