Digital twins in healthcare have predominantly focused on object-level applications, including personalized medicine, organ twins, and wearable-based patient monitoring, whereas system-level digital twins that support the planning and operation of healthcare facilities remain largely unexplored. This study addresses this gap by proposing a revised typology of healthcare digital twins and developing a system-level digital twin for hospital operations. The proposed typology integrates the healthcare facility life cycle with an analytics framework that combines descriptive and diagnostic analytics while introducing a prospective analytics dimension to support simulation-based what-if analyses. The digital twin was developed for a multiservice ambulatory care center within the CIUSSS du Centre-Ouest-de-l'Île-de-Montréal, Canada, integrating blood collection, screening, medical examinations, and vaccination services. A hybrid agent-based and discrete-event simulation model was developed in AnyLogic using service-time and arrival estimates derived from field observations, managerial input, and clinical expert consultation. Six operational scenarios involving alternative staffing levels and service durations were evaluated using 500 simulation replications per scenario. The proposed typology comprises twelve categories, seven of which are represented by previously reported healthcare digital twins, while the presented application occupies the system-level design category. The baseline configuration proved operationally infeasible, with an average patient length of stay of 266 minutes and service completion extending to 01:29. Improving staffing or service durations individually was insufficient to achieve acceptable operational performance. In contrast, jointly optimizing staffing and service durations reduced the mean patient length of stay to 75 minutes and enabled completion of daily operations by 19:14, representing a 3.5-fold improvement over the baseline. These findings demonstrate that system-level digital twins can provide actionable analytics for healthcare facility planning by enabling the evaluation of operational alternatives before implementation, while the proposed typology offers a structured framework for classifying and comparing healthcare digital twins across the healthcare facility life cycle.