Predictive analytics and predictive maintenance are expected to bring about a service revolution and fundamentally change familiar business models. But what is behind these two buzzwords? And why is examining these topics worthwhile – in particular for midsize enterprises?
Predictive analytics means the collection and analysis of, for example, machine, component, and environmental data with the goal of identifying recurring patterns and preparing for events that are likely to occur. Predictive maintenance means using such data with the concrete objective of determining the ideal point in time to perform maintenance. Establishing maintenance intervals so that both failures and unnecessary maintenance are avoided is the goal. That is because costs are incurred in both cases, and these are now higher than the costs of data collection and analysis.
Predictive maintenance makes it possible to avoid unplanned downtime and minimize the time for repairs through ideal scheduling. The costs for replacement parts and tools are also lower as a rule because they can be procured far in advance. All of these aspects harbor cost savings potential of different magnitudes. Experience has shown that costs can be reduced in many areas with predictive maintenance, leading to significant overall savings.
Aside from obtaining the required IT know-how, the key challenge lies in the collection and especially the specific analysis of the data. Using scalable cloud services such as SAP Cloud Platform or the Microsoft cloud platform Azure is recommended for the implementation of predictive maintenance. Internal server infrastructures are generally not capable of handling the volume of collected data. If going to the cloud is not desirable, the security of data transmission as well as data storage must be assured in addition.
The fields of application are virtually limitless and can be illustrated best using a concrete example: The intervals for cleaning the hull of a ship. There is no doubt that the hull needs to be cleaned from time to time. But what is the ideal point in time? Analyzing the hull soiling data in conjunction with the fuel consumption data has revealed that the ideal time is much sooner than previously assumed. This is because fuel consumption increases sharply after a certain degree of soiling that used to be considered negligible. Now the ideal time for maintenance is determined individually for each ship using predictive maintenance. Preparations to clean the ship are made as soon as a certain threshold is exceeded.
You are interested predictive analytics and want to explore the possibilities in your business? Then contact our expert Torsten Kopte now.