Predictive maintenance scheduling is an optimization problem aimed at defining the best activity sequence to minimize the expected cost over a time horizon. For very-large systems such as in experimental physics, maintenance optimization turns out to be very difficult owing to analytically intractable objective functions. In this paper, a meta-heuristic predictive maintenance algorithm based on the Generalized Extremal Optimization (GEO) is presented. With respect to state-of-the-art meta-heuristic techniques, the GEObased maintenance algorithm allows optimization procedure to be configured easily through only one parameter without a numerous population. Preliminary results of the algorithm performance validation on the liquid helium storage system of the Large Hadron Collider at CERN are reported.
Generalized extremal optimization of predictive maintenance to enhance monitoring of large experimental systems
Maisto D;
2014
Abstract
Predictive maintenance scheduling is an optimization problem aimed at defining the best activity sequence to minimize the expected cost over a time horizon. For very-large systems such as in experimental physics, maintenance optimization turns out to be very difficult owing to analytically intractable objective functions. In this paper, a meta-heuristic predictive maintenance algorithm based on the Generalized Extremal Optimization (GEO) is presented. With respect to state-of-the-art meta-heuristic techniques, the GEObased maintenance algorithm allows optimization procedure to be configured easily through only one parameter without a numerous population. Preliminary results of the algorithm performance validation on the liquid helium storage system of the Large Hadron Collider at CERN are reported.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.