No abstract

Predictive Maintenance has gained a lot of attention in recent years due to the development of improved sensors and intelligent algorithms. These allow for monitoring the health condition of production machinery and predict its future deterioration. In order to generate added value for industrial use cases, two more steps are required: considering the machine's time-varying operational conditions and integrating its dependent deterioration prediction in a holistic scheduling approach. This publication identifies a shortage of deterioration estimation frameworks under time-varying operational conditions as well as a lack of Predictive Maintenance integrated scheduling problems in the literature. Subsequently, a new conceptual framework to model future machine deterioration under time-varying operational conditions and its application in production scheduling is introduced. The Operation Specific Stress Equivalent (OSSE) represents the load of a future production job on the machine and supports a general formulation of the maintenance integrated job shop scheduling problem (MIJSSP). This formulation is presented together with benchmark instances and corresponding sample data. Finally, the formulation is tested with the help of a genetic algorithm that illustrates the potential of using new objective functions for decision support, such as the Reliability Weighted Makespan Cmax R .

Monitoring the electric power consumption of machines in manufacturing becomes more and more important due to rising electricity costs and environmental awareness. This paper proposes a cost-efficient monitoring concept requiring significantly fewer electricity meters. Power consumption of several machines is monitored aggregately and then broken down into the estimated power consumption of each machine by means of a disaggregation procedure utilizing machine data acquisition. In this paper the disaggregation procedure is presented, uncertainty limitations are discussed and an application example of an industrial environment is given.

No abstract

*) Danksagung Die Autoren dieses Beitrags bedanken sich bei der Europäischen Kommission für die Förderung des Projekts "PreCoM-Predictive Cognitive Maintenance Decision Support System" unter der Grant Agreement No. 768575. Bild 1. Instandhaltung im Kontext der industriellen Revolutionen (i. A. an [4, 5])