Maintenance is a critical component of Facilities Management (FM), and with the proliferation of big data, Internet of Things (IoT), and Industry 4.0, predictive maintenance (PdM) has emerged as a critical maintenance technique. However, modern data-driven PdM tactics are based on sensor data, but there is no obvious way to imply PdM on older buildings that lack sensors. EQUANS is a company seeking recommendations for implying PdM in the management of a historic building. This paper demonstrates the potential of survival analysis with data-driven PdM using EQUANS's non-sensored data, explicitly using the Kaplan-Meier method, parametric methods, Cox proportional hazard model, and accelerated failure time models. The boiler was chosen as the asset to focus on in this project, and the results indicated that the boiler's survival might not be related to the frequency of service but the boiler's age. The research findings propose a further step toward PdM for assets without sensors, and data collection and preventive maintenance can be improved.