There is a lot of talk about Condition Monitoring and Predictive Maintenance these days. And the key objective is always to improve and shorten maintenance cycles such that all activities will be carried out exactly at the right point in time. Such a fully digitized type of maintenance is certainly the ideal case; however, it requires a total networking of all machines and technical facilities. Under a technical point of view, yet a lot is already possible. Think about upgradeable sensors, the optimization of existing automation systems or in other ways.
In the scope of all these important considerations, though, you should never forget to take into account the expertise and experience of your staff on site. The “expert eye” of an experienced technician provides already today a precise overview about the condition of technical facilities. The art of management consists in the capturing and subsequent analysis of this valuable knowledge.
It is exactly at this point where an extension of speedikon® C Maintenance comes into action: a feed-back report of the acting technician regarding the condition of the technical facility at the point in time of executing a specific measure is being captured and provides the basis for a new calculation of maintenance cycles.
Via easy to handle selection catalogs, the executing person can give a standardized feed-back on the timewise necessity of the executed measure. For measures having been taken too early according to the executing person, the respective future cycle can be prolonged. On the other hand, the future cycle can also be shortened for measures overdue.
On the system’s side, there is definition and monitoring, whether a modification of the cycle is necessary in the first place, and for which technical facility or which component. For many components specific legal or manufacturer-dependent maintenance cycles are in place, that cannot be adjusted at all, or only under clearly determined circumstances. Concerning all other technical facilities, it is at the discretion of those responsible to find a respective balance between an increasing downtime risk on one hand and reducing maintenance costs on the other.
Whereas future cost statements are been calculated in speedikon® C these days already, we are currently examining in our research lab AI methods to determine the probability of default. The objective of this research is that an algorithm determines the optimal maintenance cycle based on the respective customer’s technical facilities’ portfolio and its individual operation to minimize both downtime risk and maintenance costs.
Here we are back at the beginning again: it is only us to realize Predictive Maintenance without installing compellingly a lot of sensors.
In case you take further interest in this topic, or even want to participate in the research activities mentioned above, please contact us at your earliest convenience.