The Predictive Maintenance (PdM) Interest Group aims to initiate the development of a common framework for applying PdM to scientific infrastructures, facilitating the sharing of data, methods, and resources.
PdM distinguishes itself from traditional maintenance by moving beyond fixed-interval scheduled interventions (preventive maintenance) or responding only after a failure occurs (corrective maintenance). Instead, it utilizes data-driven models to forecast potential equipment malfunctions in advance. This enables timely intervention before actual failures occur, minimizing downtime and the costs typically associated with corrective repairs.
Objectives
Collaborations on Joint Projects: To foster partnerships among members and participate in collective research initiatives.
Establishment of a Common Framework: To create a shared structure that streamlines the application of PdM across scientific infrastructures, promoting the exchange of data, methodologies, and resources within INAF (National Institute for Astrophysics).
Technological Exploration: The group aims to explore the use of anomaly detection and Machine Learning models applied to sensor data and log files. The goal is to identify irregular patterns and develop PdM applications that can be shared throughout the collaboration.
Group Members
Federico Incardona – federico.incardona@inaf.it
Alessandro Costa – alessandro.costa@inaf.it
Pietro Bruno – pietro.bruno@inaf.it
Vito Conforti – vito.conforti@inaf.it
Farida Farsian – farida.farsian@inaf.it
Francesco Franchina -francesco.franchina@inaf.it
Stefano Germani – stefano.germani@inaf.it
Fulvio Gianotti – fulvio.gianotti@inaf.it
Alessandro Grillo – alessandro.grillo@inaf.it
Giuseppe Leto – giuseppe.leto@inaf.it
Kevin Munari – kevin.munari@inaf.it
Francesco Schillirò – francesco.schilliro@inaf.it
Gino Tosti – gino.tosti@inaf.it