Production downtime - the worst-case scenario for every series producer. Just five minutes of downtime in automotive production is associated with average costs of 100,000 euros. To prevent such downtimes, machines, systems or means of transportation must be serviced in good time before critical components fail and production losses occur. A few years and decades ago, this only seemed possible in science fiction films. Thanks to improved and cheaper sensor, transmission and data storage technology, the predictive maintenance of production processes is already a reality in some sectors and shows great potential in the context of Industry 4.0.

To enable companies to independently develop and offer predictive maintenance systems and services in the field of predictive maintenance, the Corporate Development department of the Chair of Production Systems at the Laboratory for Machine Tools and Production Engineering WZL at RWTH Aachen University has developed a generic guideline for action in collaboration with various series producers and toolmaking companies. For the successful application of predictive maintenance, a close and cooperative partnership between series producers and toolmaking companies is essential in order to synergistically combine the advantages of tool and process knowledge in series production.

The guideline's target group includes, in particular, companies that are experiencing increased tool-related failures in their series production due to repeated, unforeseen disruptive influences. Predictive maintenance can help them to predict malfunctions such as tool failure and derive specific measures based on this. This gives tool manufacturers the opportunity to expand their existing service portfolio with predictive maintenance solutions in order to significantly increase customer benefits over the entire life cycle of the tool and open up additional business areas.

Procedure for company-specific implementation
The functionality of predictive maintenance is based on the collection, transmission, storage and near-real-time utilization of large volumes of data. Based on complex analysis procedures and algorithms, deviations in the recorded operating parameters of a machine-tool system can be identified and necessary maintenance can be anticipated. As both the technical implementation and the embedding of the technical solutions in the existing product and service portfolio often represent major challenges for series producers and toolmaking companies, the guideline is based on a comprehensive study which, in addition to concrete research results, is also based on the expert knowledge of the participating partners from industry.

The generic guideline presents a systematic procedure for developing predictive maintenance solutions in three phases with a total of six steps. As part of the analysis phase, all relevant prerequisites and requirements for a predictive maintenance solution are first recorded. In the design phase, these are translated into tool, infrastructure and service solutions. Finally, the implementation phase involves commissioning, teaching the algorithm and defining interaction points and workflows.

The results of the study show that the use of a predictive maintenance solution offers great potential for increasing machine availability by significantly reducing unplanned downtime while at the same time reducing maintenance costs through more predictable, condition-based maintenance in series production. By providing appropriate services, toolmaking companies have the opportunity to expand their offering, effectively differentiate themselves from the competition and increase their profitability. By cooperatively developing predictive maintenance solutions, both sides can benefit equally from corresponding service concepts.