In this paper we present PANOPTES, a digital twin ontology designed to support dynamic monitoring, predictive analytics, and decision-making in the management of cultural heritage assets. Building on standards such as CIDOC CRM, SOSA/SSN, PROV-O, GeoSPARQL, and OWL-Time, PANOPTES introduces a unified semantic model that represents assets, observations, diagnoses, predictions, threats, and decision processes. The ontology naturally maps to a relational backend, enabling efficient data ingestion, traceability, and operational deployment. Application scenarios include condition monitoring, risk forecasting, emergency response, and preventive conservation planning. PANOPTES enables structured monitoring, predictive threat modeling, and decision support for the preventive conservation of cultural heritage assets, particularly in remote and infrastructure-poor sites.