In maintenance management, a distinction is made between corrective, preventive and predictive maintenance. In corrective maintenance, machines are only repaired when they break down. Accordingly, there are no costs for maintenance and maintenance planning, but failures lead to unplanned system stops, the machine may produce significantly more rejects due to wear and tear and the life cycle of the machine is significantly reduced.
In prospective or preventive maintenance, the management of maintenance is based on the manufacturer’s specifications or empirical values for MTBF (mean time between failure). This increases planning reliability for maintenance and the life cycle of the system can be extended. However, unforeseen system stops can still occur, as maintenance is based on statistics and not on the actual condition of the system. This can also lead to a machine being serviced faster than necessary, resulting in additional costs for the company.
Predictive maintenance is condition-based and is therefore geared towards the current condition of the system. The basis for this is provided by real-time data that is collected using sensors and modern measurement technology and interpreted by an AI (machine learning solution). The more precisely the maintenance software is tailored to the system, the more efficiently maintenance can be carried out.