Nur in englischer Sprache verfügbar.
Data-driven analytics enables increasingly accurate predictions about the future. This has sparked major interest in the use of machine monitoring data to estimate when parts, machines or systems will fail; otherwise known as Predictive Maintenance.
It is not a new practice – predictive maintenance has been common in industrial manufacturing since the 1980s. However, with big data now in play, and with growing pressure to cut costs, organisations can no longer afford to ignore predictive maintenance when developing their maintenance strategies. This paper will discuss the merits of an optimised maintenance strategy based upon knowledge.
Firstly, we explore the difference between data, information and knowledge, all of which form the basis for any approach to maintenance. Secondly, we look at the three possible approaches to maintenance, depending on the individual characteristics of the selected part. Finally, based on these characteristics, we explain how to derive an effective, economical Knowledge Based Maintenance (KBM) strategy.
However, the implementation of such a strategy is not without its challenges, many of which will also be discussed.