An operational key performance indicator framework and reference data from a multi-room pencil beam scanning proton therapy center

Purpose: Multi-room proton therapy centers face unique operational challenges related to shared beam delivery systems, extended treatment hours, and high clinical complexity. This study presents a structured framework of key performance indicators (KPIs) and reports longitudinal single-institution operational reference data from a multi-room proton therapy center to support operational assessment and continuous quality improvement. Methods and materials: Operational KPIs were retrospectively evaluated for calendar year 2024 at a three-room proton therapy center utilizing pencil beam scanning. KPIs were categorized into three areas: beam performance, treatment planning and pre-treatment workflow, and treatment room operations. Beam performance metrics included site- and room-specific uptime and beam pauses. Treatment planning metrics quantified CT simulation–to–patient scheduling timelines with attribution to physician, dosimetry, and physics components. Treatment room operations metrics included appointment variance, slot utilization, and waiting-for-beam (WFB) time. Data were collected automatically using a commercial analytics platform integrating radio-frequency identification tracking and oncology information system. Metrics were analyzed longitudinally and stratified by treatment room, disease site, and workflow component. Results: For calendar year 2024, aggregate site uptime was 95.3%, weighted according to clinical operating time across the three treatment rooms; individual room uptimes ranged from 91.8% to 97.2%. Beam pauses were frequent but brief, resulting in 2,238 minutes of cumulative lost treatment time across 2024, equivalent to approximately 0.5% of total clinical operating time. Annual mean treatment planning time was 8.5 business days, with physician-related activities contributing the greatest variability across disease sites. Monthly average appointment variance ranged from 5 to 11 minutes, with most treatments initiated within 10 minutes of scheduled time. Across 2024, slot utilization was stable, with 73% of fractions completed within scheduled slots. Mean overrun and underrun durations were modest, although P90 overrun durations exceeded 10 minutes throughout the year. Mean WFB times were low and consistent, ranging from 1.5 to 2.3 minutes. Conclusions: This study presents a practical KPI framework and a single-institution operational reference data for multi-room proton therapy center. These metrics may support longitudinal monitoring, identification of workflow inefficiencies, data-driven decision-making, and future multi-institutional benchmarking efforts.