Increasing patient throughput in radiation oncology: The use of discrete event simulation computer modeling as a management tool

Robin Famiglietti, The University of Texas School of Public Health


The objective of this study was to improve patient throughput at a state-of-the-art radiation oncology center by enhancing the treatment delivery process. This improvement in clinical efficiency was achieved by developing and using a discrete event simulation (DES) computer model with which to make evidence-based operational management decisions regarding changes in processes. The study design focused on a process improvement initiative for patients receiving radiation therapy at MD Anderson Cancer Center. Baseline data for quantifying various components of the patient process from check-in to treatment completion were gathered in 2013; these data were used to develop the DES computer model and to identify clinical inefficiencies in the system. The DES computer model represents the Department of Radiation Oncology unit at the Houston main building, with nine patient tracks (representing different treatment machines) and discrete event activities specific to the patient process flow from check-in to departure. After the DES computer model was verified and validated, "what if" scenarios were developed and tested to identify possible enhancements in operational processes. Baseline operational metrics demonstrated that both machine and staff utilization was quite low (corresponding averages 58% and 56%). Closer analysis revealed inefficiencies in scheduling practices for both patients and staff. Wait time was also identified as a factor needing improvement, as it comprised 49% of the patients' total cycle time. Use of the DES computer model to break down the process data was critical to understanding the opportunity to influence the primary objective of reducing the patients' total cycle time in the radiation oncology clinic. "What if" scenarios were identified and conducted based on ideas from stakeholders. The most impactful scenario was found to be adding patient dressing rooms outside the linear accelerator, which are not currently present at the main campus. Comparing the three experiments in this study, total cycle time (36%) and wait time (23%) were reduced the most when dressing rooms were added. The effects on total cycle time and wait time were less substantial but still significant at 9% and 18% when specialization of a minimum number of treatment machines was considered. Effects on total cycle time (11%) and wait time (19%) were similar when volume modulated arc therapy was used, with an estimated reduction in process time by 6 minutes.

Subject Area

Health care management|Operations research|Oncology

Recommended Citation

Famiglietti, Robin, "Increasing patient throughput in radiation oncology: The use of discrete event simulation computer modeling as a management tool" (2014). Texas Medical Center Dissertations (via ProQuest). AAI3639714.