Visualization of Waiting Times and Evaluation of Workload Congestion Mitigation in General Imaging

Abstract

Purpose: This study aimed to visualize patient waiting times in general radiography using a data-driven approach and to evaluate operational strategies for congestion mitigation in a high-volume clinical setting. Methods: General radiography examination data collected over a 3-year period were retrospectively analyzed. Waiting time was defined as the interval between patient reception and examination start. The overall distribution and characteristics of waiting times were evaluated using inverse cumulative distribution analysis. Factors influencing waiting time were assessed using random forest regression, incorporating examination-related and operational variables. Based on the identified key factors, a discrete-event simulation model was constructed to evaluate the effects of different operational conditions, including staffing levels, on patient waiting time. Results: Inverse cumulative distribution analysis revealed substantial variability in waiting time across examinations. Random forest analysis identified examination type, imaging region, and operator as major contributing factors to waiting time. Discrete-event simulation demonstrated that increasing staffing levels effectively reduced waiting time; however, the magnitude of improvement diminished beyond a certain staffing threshold, indicating limited returns from simple workforce expansion. Conclusion: Data-driven visualization and machine learning–based analysis are effective for understanding waiting time characteristics in general radiography. The findings suggest that congestion mitigation cannot be achieved solely through staffing increases and requires optimized workflow design and operational planning based on empirical data.