Bibcode
Rueda-Teruel, F.; Yanes-Díaz, A.; Rueda-Teruel, S.; Bello, R.; Ventura-Guillamón, J.; Domínguez-Martínez, M.; Martínez-Olivar, N.; Olivas-Galve, J. M.; López-Alegre, G.; Chueca, S.; Rodríguez-García, L. A.; Domínguez, R.; de Paz, H.; Huertas, M.; Iñiguez, C.; Marin-Franch, A.; Civera, T.; Collado-Escusa, J. J.; Mayoral-Blasco, M. R.; Albero-Gamir, J.; Garcés-Cubel, D.; Farah-Amin, S.; Almarcegui-Gracia, M.; Cenarro, A. J.; Corradi, R.; Moles, M.; López-Sanjuan, C.; Vázquez-Ramió, H.; Zaragoza-Cardiel, J.; Valdivielso-Casas, L.; Lizaga-Villarroya, J.; Pyrzas, S.; Díaz-Martín, M. C.; Iglesias-Marzoa, R.; Cabanillas-de-la-Casa, C.; Maícas-Sacristán, N.; Lamadrid-Gutiérrez, J.; López-Martínez, F.; Galindo-Guil, P.; Lacruz-Calderón, E.; Castillo, J.; López-Sainz, A.; Akhlaghi, M.; Hernández-Fuertes, J.; Muniesa-Gallardo, D.; Moreno-Signes, A.; Hernán-Caballero, A.; Ederoclite, A.; Fernández-Ontiveros, J.; Infante-Sainz, R.; Eskandarlou, S.; Fernández-Gil, D.; Hidalgo-Pinilla, A.; Morate-González, D.; Acharya, N.; Anguiano, B.; Arellano-Cordova, K. Z.; Kulkarni, S.; Martínez-Delgado, D.; Novakovic, B.; Sáez-Ruiz, F. J.; Zarattini, S.; Vaquero-Valer, S.; Muñoz-Igado, I.; Alegre-Sánchez, M.; Julián-Caballero-de-España, G.; Allueva-Gómez, R.; López-Colás, A.
Bibliographical reference
Ground-based and Airborne Telescopes XI
Advertised on:
8
2026
Citations
0
Refereed citations
0
Description
Artificial intelligence (AI) is poised to transform maintenance in astronomical observatories by moving from purely corrective and preventive routines towards predictive and condition-based strategies. We present OPTIMIA, an AI-enabled framework at the Javalambre Astrophysical Observatory (OAJ) that fuses sensor data, maintenance records, and a virtual assistant to support technicians and maintenance managers. Initial targets include diesel generators, glycol-water pumping groups, HVAC plants, electrical power distribution, and liquid-nitrogen supply. Using machine learning on this heterogeneous data, OPTIMIA detects early signs of degradation, proposes intervention windows, and prioritises work orders according to risk and scientific impact. The virtual maintenance agent delivers these capabilities through natural-language recommendations in the existing control and maintenance environment, reducing unplanned downtime and enabling data-driven decisions for observatory infrastructure.